This Week in Tech Episode 1086 Transcript
Please be advised that this transcript is AI-generated and may not be word-for-word. Time codes refer to the approximate times in the ad-free version of the show.
Leo Laporte [00:00:00]:
It's time for TWiT this Week in Tech. Great panel for you. Pulitzer Prize winning author Gary Rivlin is here. The author of AI Valley, Molly White from Web3 is going great. And Sam Abulsamet, my car guy. We will talk about cars, the brand new Luce from Ferrari. Sam has opinions. Google's changing its search page.
Leo Laporte [00:00:20]:
Molly's mom has opinions. And what did the Pope say about AI? And why is Gary in agreement? All of that and more coming up on. Stay tuned.
Gary Rivlin [00:00:32]:
Podcasts you love from people you trust.
Leo Laporte [00:00:37]:
This is twit. This is TWiT this Week at Tech. Episode 1086, recorded Sunday, May 31, 2026. The great beagle migration. It's time for TWiT this Week in Tech, the show where we cover the week's tech news. There's quite a bit of tech news this week and good news. We have a fabulous panel to cover it. Sam Abulsamet is here.
Leo Laporte [00:01:06]:
My car guy from Wheel Bearings media and the VP of research@telemetryagency.com. hi, Sam.
Sam Abuelsamid [00:01:13]:
Hi, Leo. How are you today?
Leo Laporte [00:01:14]:
I am very, very well, thank you very much. And it's always nice to see you. From Ypsilanti, Michigan.
Sam Abuelsamid [00:01:22]:
Yeah.
Leo Laporte [00:01:23]:
Home of Ford. Of you. Of you.
Sam Abuelsamid [00:01:28]:
Ford's not here.
Leo Laporte [00:01:29]:
Ford's not near Fort. Down the road in Dearborn.
Sam Abuelsamid [00:01:32]:
Yeah.
Leo Laporte [00:01:32]:
Nearby. Molly White's also here. World renowned Wikipedia editor and the creator of Web3 is going just great. And her newsletter is Citationneeded News. Hi, Molly. So how is Web three going?
Molly White [00:01:51]:
Just great.
Leo Laporte [00:01:55]:
There's actually some funny bitcoin news we'll talk about in just a little bit.
Sam Abuelsamid [00:01:59]:
Has it fallen over yet?
Molly White [00:02:02]:
I would say it fell over quite a while ago and it just sort of continues. It's like it's going down the stairs or something. Continuing to fall.
Leo Laporte [00:02:09]:
You don't really need to. It's not even a question anymore, is it? You know, it's like, it's like if you talk about NFTs now, nobody goes, oh, these are the greatest thing ever.
Molly White [00:02:19]:
Yeah. I think people have mostly ditched the Web three terminology and are just sort of going, you know, still trying to convince people that maybe crypto is a thing or bitcoin is still a good thing. But you don't hear much about Web3 anymore.
Leo Laporte [00:02:31]:
No, but you cover NFTs and crypto on Web3 is going just great. It's all in. All of. All of a bundle thing.
Molly White [00:02:40]:
Yes. And it's all in bitcoin, much to the chagrin of some bitcoiners. Who are very upset that I.
Leo Laporte [00:02:46]:
Well, you keep under that umbrella, you keep. You know, man, we're, we're like, we're way down, man. It's all Molly White's fault. That's the. I know that's the truth. Also with us, for me, we'd be at a million. If it weren't for you, bitcoin would be worth a million dollars. That's what.
Leo Laporte [00:03:00]:
That's right. That's right. Gary Rivlin's also here, a Pulitzer Prize winning journalist. His most recent book, AI Valley covers Microsoft, Google and the trillion dollar race to cash in on artificial intelligence. It's still true. The book came out a year ago now.
Gary Rivlin [00:03:19]:
Yeah, like 14, 15 months ago.
Leo Laporte [00:03:21]:
Yeah. And yet here we are with the IPOs for OpenAI around the corner and Anthropic around the corner. And I mean, the trillion dollar race has not yet won.
Gary Rivlin [00:03:34]:
Yeah, you look at the S1 for SpaceX.
Leo Laporte [00:03:39]:
SpaceX is wild.
Gary Rivlin [00:03:40]:
I mean, I mean I actually have. I just think it's so.
Sam Abuelsamid [00:03:43]:
How could anybody read that document, take this company seriously?
Leo Laporte [00:03:48]:
There's so many juicy tidbits in there. It came out while we were doing intelligent machines on, on Wednesday last week. And Paris, like we lost her because she was like, oh my God. And then there's this, and then there's this. We, for instance, X. We're learning about X.com and how little money they're making. What did you notice?
Gary Rivlin [00:04:06]:
Making their. How much money they're losing.
Leo Laporte [00:04:08]:
Losing.
Gary Rivlin [00:04:09]:
He talks about the addressable, actionable, total. Excuse me, the actual total addressable market, including 26.5 million. I love the 0.5 trillion to be made in AI and it just.
Leo Laporte [00:04:22]:
Yeah, it's made up.
Gary Rivlin [00:04:24]:
I really don't. It's a made up number. And you know, they are in a race with a lot of other competitors. It just, it just, it was so musky and stuff.
Sam Abuelsamid [00:04:32]:
Actually my favorite, my favorite part for the course for Elon, it is. It's all made up.
Gary Rivlin [00:04:38]:
Well, so you know, he famously had his, his pay package that if he could create a colony at Mars.
Leo Laporte [00:04:47]:
A million people on Mars.
Gary Rivlin [00:04:49]:
People on Mars and data centers in space get this gazillion number of shares,
Leo Laporte [00:04:54]:
there'll be a trillionaire. Yeah, yeah.
Gary Rivlin [00:04:55]:
And it works out that in the S1, even though he's not granted the shares until he reaches these crazy benchmarks, he still gets the votes.
Leo Laporte [00:05:06]:
Oh yeah, 10 to 1, right? Something like that.
Gary Rivlin [00:05:09]:
Well, not just 10 to 1, but on shares he does not own until he hits these benchmarks and yet, the way the S1 is written, he could still vote on the shares he doesn't yet have. He could still vote using the shares he doesn't have yet.
Leo Laporte [00:05:23]:
So. And yet people are probably going to race to buy SpaceX stock when it goes public, right?
Gary Rivlin [00:05:32]:
I mean, everyone wants. Or not everyone, lots of people want a piece of AI. The weird thing about SpaceX is Starlink's probably their best product, their most profitable product. And so much of that company is wound up in xai. It's why they're losing so much money, losing tens of billions of dollars to xai. And it's called Space. It's like Meta. Is it Meta? But it still has that name.
Gary Rivlin [00:05:58]:
But people want to own a piece of anything that lets them own a piece of AI.
Leo Laporte [00:06:03]:
Right.
Molly White [00:06:04]:
There's also been a meme stock for years. Right. So people probably assume the same thing will happen with SpaceX.
Leo Laporte [00:06:12]:
The big difference is SpaceX is rolling in federal subsidies. In fact, they just signed another.
Sam Abuelsamid [00:06:17]:
But they still lose money if you take out Starlink. The actual space business of SpaceX, of launching stuff, loses money.
Leo Laporte [00:06:26]:
Right.
Sam Abuelsamid [00:06:27]:
So even though they made this amazing breakthrough of reusable rockets that were supposed to make the cost of launches so much cheaper, they still lose money on it.
Leo Laporte [00:06:40]:
Well, but you can afford to lose money when you have an $8 billion federal contract, right? I mean, we. It's really the taxpayers who are losing the money, right?
Sam Abuelsamid [00:06:51]:
Well, no, I mean, we, you know, we're paying, but we're paying less than what it costs SpaceX to do those launches.
Leo Laporte [00:06:59]:
So they still lose money even with the federal subsidies.
Sam Abuelsamid [00:07:03]:
Yeah, I mean, they're technically not getting subsidies. That's. That's business.
Leo Laporte [00:07:07]:
I mean, yes, you could $38 billion in business.
Sam Abuelsamid [00:07:11]:
Yeah, and, and, and they spend more than that on actually doing the launches.
Leo Laporte [00:07:16]:
That's kind of pathetic.
Gary Rivlin [00:07:20]:
Least enough Blue Origin this week.
Leo Laporte [00:07:22]:
Oh, man. Well, I guess we're in our space segment. Might as well keep going on that. I was gonna start with the Pope, but let's talk about Rocket go Boom. This is the most spectacular rocket explosion since N1 just happened in Florida. This is the new Glenn. And they were just doing a loading. They weren't even launching it.
Leo Laporte [00:07:44]:
Right. Or maybe they were.
Sam Abuelsamid [00:07:47]:
They were doing a hot fire test.
Leo Laporte [00:07:48]:
Hot fire.
Sam Abuelsamid [00:07:49]:
So they, they load up the fuel, you know, fire the rockets or fire the engines without actually launching the vehicle. So it's still clamped to the launch pad.
Leo Laporte [00:07:59]:
So it wasn't supposed to go in the air, but it kind of sort of did.
Sam Abuelsamid [00:08:03]:
Pieces kind of went everywhere, every direction.
Molly White [00:08:05]:
Explosion like that is going to go somewhere.
Leo Laporte [00:08:08]:
Just damaged the pad significantly. It's the only pad.
Sam Abuelsamid [00:08:11]:
I would say more like destroyed the pad.
Leo Laporte [00:08:13]:
Yeah, that's the one new or only one Blue Origin has, I think.
Sam Abuelsamid [00:08:18]:
Yeah. The only one for new Glenn. Yeah, yeah.
Gary Rivlin [00:08:21]:
And I think, you know, kind of the bigger picture here is, you know, NASA, we, whatever, were counting on Blue Origin keeping SpaceX honest. Right. A competitor and, you know, they are very far behind.
Leo Laporte [00:08:33]:
All of a sudden the N1 explosion was in 1969. That was a Soviet rocket that was destroyed in a launch attempt. That's how long it's been. But, you know, we really kind of started launching a lot more rockets and Elon has been doing pretty well with his Falcon Heavy and so forth. So what was the new Glenn gonna do on the lunar landing? What part of it was it?
Sam Abuelsamid [00:08:57]:
It was gonna launch the lander, the human lander.
Leo Laporte [00:09:01]:
Okay, so much for that.
Sam Abuelsamid [00:09:02]:
Actually, the first thing that was supposed to go up was the rover. So they were gonna launch a lander that had the new Lunar RO2 rovers.
Leo Laporte [00:09:13]:
They're gonna launch.
Sam Abuelsamid [00:09:14]:
Yeah, right. And that was supposed to go up next. Next year sometime. But it's going to take them a minimum of a year to rebuild that launch complex.
Leo Laporte [00:09:27]:
Wow.
Gary Rivlin [00:09:29]:
Now, SpaceX had some. Yeah, SpaceX had some famous kablooies, but. Oh yeah, they were.
Sam Abuelsamid [00:09:34]:
Lots of them.
Gary Rivlin [00:09:35]:
They were in the air. Right. I mean, it's never on the launch pad. Is that what makes this so unique?
Sam Abuelsamid [00:09:39]:
They've had them. They've had some that blew up on the launch pad. There was a starship booster about three years ago, I think. Three, four years. It was one of the first starship attempted launches. And it blew up on the pad and destroyed much of the pad as well.
Leo Laporte [00:09:58]:
So, I mean, in a way, this we kind of forget because they've had some really successful missions. The most recent Artemis 2 around the moon. And you forget how dangerous and iffy all of this is. This is not easy to do and crazy expensive. And it's crazy expensive. I mean, Jeff Bezos, who funds Blue Origin, can afford to rebuild the pad, but it does slow down NASA's plan to get to the moon, right?
Sam Abuelsamid [00:10:31]:
Yeah, I think the, the diff. You know, the fundamental difference is that since, particularly since the Apollo 1 disaster, when they had a fire and in the. The capsule on Apollo 1 before the launch and the three astronauts in there were killed after that happened, you know, it fundamentally changed the way NASA operated. And you know, they. They have still had issues since Then you know, obviously Apollo 13 had problems. They, we had two space shuttles blew ups. Yeah. But you know, they take, they spend a lot more time up front testing stuff, you know, and so they've had a much lower percentage of, of, of explosions, you know, either at launch or in transit than SpaceX has had.
Sam Abuelsamid [00:11:26]:
SpaceX, you know, it's move fast and break things kind of philosophy. Whereas NASA takes the time, you know, it, it takes them a lot longer to get there. But you know, they have, you know, they've sent Artemis, you know, they've sent the, the space launch system around the moon now and back safely with astronauts.
Molly White [00:11:47]:
Well, the tech industry is just bringing move fast and break things to like.
Leo Laporte [00:11:51]:
Right.
Molly White [00:11:51]:
As many industries as possible where you would never want that to be the philosophy.
Gary Rivlin [00:11:55]:
Including the US Government and world, World organizations. Yes.
Molly White [00:12:01]:
Yeah.
Leo Laporte [00:12:01]:
Well, China just had a rocket explosion too. Right. So it's just, this is, we forget how dangerous and difficult this is.
Sam Abuelsamid [00:12:10]:
Well, you know, one, and one of the things that's in the SpaceX IPO somewhere, they talk about terrestrial launches, you know, point to point. Terrestrial.
Leo Laporte [00:12:20]:
That's what they travel for cargo or for humans.
Sam Abuelsamid [00:12:23]:
For humans.
Leo Laporte [00:12:24]:
So like you could fly to, you could fly from New York to Australia.
Sam Abuelsamid [00:12:27]:
Yeah. Basically anywhere on the planet in half an hour. Yeah, yeah. Given how often these things blow up, do you really need to get to Australia in half an hour? I sure as hell don't, you know, and then there's also the environmental impact. Yeah. Because these things are burning huge amounts of methane.
Leo Laporte [00:12:47]:
You know, these, this was a methane.
Sam Abuelsamid [00:12:49]:
Yeah. All these new rockets, all, all the SpaceX and Blue Origin, most of these new rocket companies are all doing methane rocket engines. So they're burning enormous amounts of methane. So the, the CO2 emissions from this, not to mention, you know, just the methane that's vented during fueling process, far
Leo Laporte [00:13:08]:
worse greenhouse gas than almost anything.
Sam Abuelsamid [00:13:10]:
And you know, you're talking about filling these things with liquid methane. It's a, you know, because it's a liquid, you know, they're in cryogenic containers rather than in pressure. High pressure. High pressure containers, which means that, you know, this stuff is boiling off continuously. If you, if you ever watch one of these things on the pad, you'll see that the clouds streaming off these things as it boils as the, the methane.
Leo Laporte [00:13:34]:
That's methane.
Sam Abuelsamid [00:13:34]:
Liquid oxygen. Yeah, boils off.
Leo Laporte [00:13:37]:
And so Elon loves his methane. He's using it to power his data centers as well.
Sam Abuelsamid [00:13:42]:
Yeah.
Gary Rivlin [00:13:42]:
Doesn't he use it at night too? Oh, no, that's Something El. Nevermind.
Leo Laporte [00:13:45]:
No, that's kethamine.
Molly White [00:13:47]:
Methane in the morning.
Gary Rivlin [00:13:48]:
Ketamine.
Leo Laporte [00:13:49]:
Kethamine. Oh, I shouldn't joke about it.
Gary Rivlin [00:13:53]:
Oh, why not?
Leo Laporte [00:13:55]:
Do they mention ketamine and risk factors in the SpaceX IPO?
Gary Rivlin [00:14:00]:
I did not read the whole thing. I read the articles. It's also interesting. There's this standard thing that people, you know, that publicly traded companies do like, you have a number of independent directors. You only do compensation through a sub committee of independent directors. And they just toss that out the window.
Leo Laporte [00:14:20]:
Right.
Gary Rivlin [00:14:21]:
To. They're not following that practice of having
Leo Laporte [00:14:23]:
clearly Elon's company from start to finish. Right. He has.
Sam Abuelsamid [00:14:27]:
Yeah. There's. There's no real corporate governance in any of Elon's enterprises.
Leo Laporte [00:14:31]:
Right. Because it's Elon.
Gary Rivlin [00:14:32]:
Well, I mean, in defense, you know it. Go ahead.
Molly White [00:14:36]:
The SEC is also a lot less, I think.
Leo Laporte [00:14:39]:
Right.
Molly White [00:14:40]:
Interested in those types of things these days. And so perhaps not applying the same level of scrutiny to these filings.
Leo Laporte [00:14:47]:
Well, and let's not forget the irony of Elon trying to save federal money with Doge claiming to try to. Claiming didn't save much, may even cost money while he's getting $38 billion in federal subsidies. There's a certain irony in all of this.
Gary Rivlin [00:15:09]:
Yeah. He got his money's worth from his $250 million. Whatever the final number was in that was. I mean, and you know, all the dropped investigations into this practice or that practice. So I guess in defensive, Musk here is like eyes wide open. No one is going to put money in SpaceX and think like, well, at least we're going to get good governance. You're basically betting on Elon Musk.
Leo Laporte [00:15:32]:
I don't know. You know what I think you're really betting on? You're betting that others will bet on it. It's like a meme. It's a meme stock. You're betting that it will go up before it goes boom. Like the rocket, like Tesla. And you know what? Tesla's still strong. Right.
Leo Laporte [00:15:49]:
Even Sam. I mean, it's still a strong stock, I think, isn't it?
Sam Abuelsamid [00:15:53]:
Yes, inexplicably.
Leo Laporte [00:15:55]:
Because it's not. The company's not has suffered a lot ever since Elon took that big political stance.
Sam Abuelsamid [00:16:01]:
I mean, if you look at the business itself, you know what their revenues and profits are. Their share price is absurd. It has no connection at all to the actual business. The share price is based almost entirely on people's belief that they're suddenly going to flip a switch someday by the end of 2020 and turn millions of Tesla vehicles into robo taxis that are generating trillions of dollars in free revenue. And that was going to happen by the end of 2020. And I believe we may have passed that date. I'm not sure what day it is right now, but I think we missed that deadline.
Leo Laporte [00:16:42]:
Yeah.
Gary Rivlin [00:16:43]:
Well, there's also the argument that they create the technology that other car makers theoretically would want. The BMWs, VWs of the world. But I happen to have written an article on this like six months or so ago, and none of the big car companies trust Tesla. There's a couple of startups out there that are selling, you know, you could get the autonomous vehicle technology through them. So even what in my mind would have been one of their big pluses? Like, okay, well we have this amazing technology now. Let's license it to other car dealers, car makers. No carmaker seems to want to trust Elon Musk or Tesla.
Leo Laporte [00:17:22]:
They're inspired by it though, right? I mean, I hear people talking about how much less wiring is in modern vehicles, and a lot of that is because they looked at Tesla and said, wow, they really were able to cut down the wire. Wiring's a. You've taught me this, Sam. There's miles of wiring in a typical vehicle and it weighs a lot, you
Sam Abuelsamid [00:17:43]:
know, copper, and copper's getting more expensive all the time.
Leo Laporte [00:17:46]:
Yeah. So people are inspired by him. Have they. But has Tesla found many licenses with car manufacturers?
Sam Abuelsamid [00:17:53]:
No, because the problem is most of what they've done is not patentable. You know, you can't license it because
Leo Laporte [00:18:01]:
they just copy you.
Sam Abuelsamid [00:18:02]:
Yeah. You know, so I mean, most, you know, many of the Chinese automakers, you know, they just straight up copied it. Ford is copying it now, Rivian's copying it, Lucid's copying it. GM is starting to copy some of that stuff. So, yeah, there's no, the problem Tesla has is most of what they're doing has no moat.
Leo Laporte [00:18:22]:
Right. They, as I remember, were one of the first to do the metal forming that they do to make the car bodies.
Sam Abuelsamid [00:18:31]:
Right. Not, not so much that they're, they were one of the first to do large scale castings for big parts of the structure.
Leo Laporte [00:18:40]:
Right.
Sam Abuelsamid [00:18:40]:
Yeah. Instead of having 100, you know, small stamped steel pieces and welding them all together, they just turned into one large aluminum casting.
Leo Laporte [00:18:50]:
Right.
Sam Abuelsamid [00:18:50]:
And again, you know, that's something that, you know, they, Tesla didn't invent that.
Leo Laporte [00:18:54]:
Patent it.
Sam Abuelsamid [00:18:55]:
Yeah, yeah. I mean, you know, they've been doing large scale castings for a long time. They just, they Were the first ones to use it in that particular application.
Leo Laporte [00:19:02]:
Were they the first to use these big industrial robots? I remember going when I bought my Model X, taking a tour of the Fremont plant, and they had these giant robots. One of the things they did and I. This seemed to me probably not the most efficient was they would work on the car upside down, and then a robot would come and pick it up and turn it on the other side as it. And then it put it on the assembly line as it continued down the line. That was impressive.
Sam Abuelsamid [00:19:33]:
Those robots have been in use since the mid-1980s.
Leo Laporte [00:19:35]:
Oh, okay. So that wasn't new.
Sam Abuelsamid [00:19:37]:
And, you know, flipping cars over, you know, to give easier, you know, provide better ergonomics for the people bolting parts on. Again, that's also something that's been done for decades.
Leo Laporte [00:19:47]:
Oh, okay. You know what else has been done for decades? Racism on the assembly line.
Sam Abuelsamid [00:19:53]:
And Nicole. That was big on the Model T line back in the 1910s.
Leo Laporte [00:19:59]:
Henry was noticeably racist.
Sam Abuelsamid [00:20:02]:
I guess Elon Musk is more like Henry Ford in some of the worst ways possible.
Leo Laporte [00:20:08]:
The California court said that's going to go ahead, that lawsuit against Tesla for racism. So Elon was trying to get that thrown out. He did win in court, though, against Sam Altman. No, no, he didn't. Never mind.
Molly White [00:20:30]:
You don't want anyone to win those.
Leo Laporte [00:20:32]:
Yeah, no, nobody wins those kinds of losses. That would be cool. Yeah. All right, well, I wasn't going to start the show with all that, but we did. And I think it's interesting. The. That rocket explosion was certainly dramatic.
Sam Abuelsamid [00:20:46]:
I remember it was spectacular.
Leo Laporte [00:20:47]:
Yeah. Saw a lot of ring camera videos from people miles away as the sky lit up. It was interesting. Yeah, we see things from many angles these days. All right, well, let's take a little break and then we'll talk, because I know, Gary, you wanted to talk a little bit about this, about Pope Leo and his encyclical on AI, which I think, on the face of it, is kind of an interesting juxtaposition of a church as essentially a medieval institution weighing in on the most modern possible technologies.
Gary Rivlin [00:21:26]:
Well, and the religion of AI. I mean, you talk to some religious folks, they talk about AGI. It does feel very religious.
Leo Laporte [00:21:33]:
Yeah. What. We'll talk about that. We will talk about the new. Actually, I'm curious what Sam thinks of the new Luce. The Ferrari. Or is it Lucci Luce, because it's Italian Luce, designed by Jony.
Sam Abuelsamid [00:21:47]:
I've.
Leo Laporte [00:21:49]:
It has switches. Ironically, Johnny was famous for not liking buttons, at least when it was Apple. And a lot more, including how Google is changing search dramatically. Plus Wikipedia editors threaten to go on strike. And Amali's going to explain what that one's all about in just a little bit. Oh, and the little Easter egg somebody buried in the blockchain.
Molly White [00:22:15]:
One of many.
Leo Laporte [00:22:16]:
One of many. By the way. It's beautiful because it never goes away. It lives forever in the blockchain. And you will be downloading it the next time you check your wallet. Let's take a little break. You're watching this Week in Tech. Gary Rivlin is here.
Leo Laporte [00:22:32]:
His book AI Valley I read and really enjoyed. And it's interesting because it really is completely timely. Last time we talked, you were going to work on another book about AI or how's that?
Gary Rivlin [00:22:44]:
I was looking at AI and energy, but I felt like I was really early on that the data centers, how it's going to overtax the grid and all this stuff. But I wrote the proposal. By the time I was passing it around, my agent was passing around. Everybody knew that story. I mean, I'll give the mainstream media credit. The Times, Wall Street Journal, Washington Post, et cetera, the world, they've really been on it. So it's one of those things like, wow, this is underreported story that I thought of about a year ago and now I wouldn't call it over reported. I think it's appropriately reported.
Gary Rivlin [00:23:17]:
So still thinking?
Leo Laporte [00:23:19]:
Yeah, I mean, if you're writing about AI, it's hard to stay ahead. I've never seen anything in technology move so fast. And I've been covering technology for three or four decades now. And this is the fastest moving story ever. We could do 50 AI stories a day if we wanted to.
Gary Rivlin [00:23:38]:
What percentage would be interesting? But anyway, I know what you're saying. Like two years ago, Google was fumbling all over the place. Now I would say Google's leading the way on AI for the moment. Anthropic was kind of the also round to OpenAI. Now it's in every way, whether it's paper value through, through investments or how its models are doing on the leaderboard.
Leo Laporte [00:24:03]:
They're profit anthropics pulled ahead. Ironically, they released a new model this week. That's crap, but we'll talk about that too.
Gary Rivlin [00:24:14]:
Well, they have the enterprise. They figured out that the way to pay for all this research is to get big businesses and they've done a great job of that now OpenAI scrambling. But my point is like a year ago it was open AI and now anthropics. So you are right. It is constantly, constantly changing.
Leo Laporte [00:24:33]:
No one's on top. This is a horse race where the lead changes every two seconds. It's just bizarre. It's just wild. Yeah. I wouldn't want to try to write a book, pick something that's not going to change. Maybe the Catholic Church would be a good subject. Something that's not going to change for a few hundred years.
Leo Laporte [00:24:48]:
That would be. That would be the topic.
Sam Abuelsamid [00:24:51]:
How about the way that some of these AI infrastructure companies are influencing politicians to get approval for data center locations where people living there don't want them?
Leo Laporte [00:25:05]:
It's fascinating to overwhelmingly unpopular.
Molly White [00:25:09]:
Gallup just did a great poll on it.
Leo Laporte [00:25:11]:
They said 71% of Americans do not want a data center anywhere near that.
Gary Rivlin [00:25:14]:
Yeah.
Sam Abuelsamid [00:25:14]:
I mean, there are. There are two that are trying to build within a few miles of me here and many more around Michigan. And Michigan Governor Gretchen Whitmer has been hugely supportive, really pushing these things, and nobody in any of these municipalities wants them anywhere near them.
Leo Laporte [00:25:34]:
Right.
Gary Rivlin [00:25:34]:
So the question I'm trying to figure out on that is, is it because, like, it's a noisy neighbor, the what's it going to mean for the grid and your electricity and your electricity prices, or is it just simply a physical manifestation of AI and we don't like it. And so we oppose data centers because it's really the only way we can express our dislike with AI Short of booing someone during a commencement speech, I
Sam Abuelsamid [00:26:01]:
would say it's actually all of the above.
Molly White [00:26:03]:
Yeah.
Sam Abuelsamid [00:26:03]:
I'll add one all of that.
Leo Laporte [00:26:05]:
Utility prices, the cost of electricity skyrocketed for a variety of reasons, I suppose. But I think people blame data centers for that and they may be right.
Gary Rivlin [00:26:14]:
Well, they are. It was flat for years. It was flat for years. In the last few years, it's gone up. I had the stat in my head a couple of months ago when I was doing this, but it's gone up 25% ish. Over the last four or five years. It's gone up a lot.
Leo Laporte [00:26:27]:
James Carville, who said it in 1992. And it's still true. It's the economy is stupid. People vote their pocketbooks, and a politician who ignores that does so at their own peril, I'm sure. Gretchen Whitmer, Governor Whitmer, supports data centers because it's good for the economy. Right. It's good for the Michigan economy. It's good.
Molly White [00:26:47]:
I think a lot of the opposition also comes from the fact that the data center, you know, the people who are pushing to build the data centers Try to position it as good for the local economy because it will create jobs.
Leo Laporte [00:26:58]:
Exactly.
Molly White [00:26:58]:
But there's actually not that many jobs involved. You know, there's this construction period which involves jobs. But then, you know, once you have an established data center, it's kind of a skeleton career, just keeping things going. And so people sort of see through that argument and they say, why are we subsidizing this? You know, why are we welcoming this when it's not bringing jobs? It's not helpful for the economy locally and it's introducing noise pollution, pollution, environmental pollution in some cases, and so on.
Leo Laporte [00:27:28]:
Here's a. This is a datacentertracker.org website where they track data center build outs, community response and legislative action. And it is a very volatile situation right now with data centers. I mean, there's just no question about it. Places that they've, they've been banned, places that they're going ahead, moratoriums.
Molly White [00:27:57]:
And now we're seeing the AI industry building these lobbying arms as well to try to get basically preemption over the state level regulations and in some cases the moratoriums against building this. And it's sort of the crypto industry playbook of dumping hundreds of millions of dollars into politics to try to find congresspeople who are willing to push for the pro A, pro AI legislation is really horrifying.
Gary Rivlin [00:28:29]:
I mean, oppose us and we'll spend millions to defeat you. Support us, be out front and you know, you be one of our advocates and we'll give you millions, tens of millions to make sure you get reelected, get elected.
Molly White [00:28:42]:
Yeah, it's literally the same strategy. I'm actually, I have a project called Follow the Crypto where I follow the crypto election spending and I've actually, I'm releasing it, I think this coming week. But I've done a huge overhaul to incorporate the AI spending because they're so similar that you can't talk about one without talking about the other.
Leo Laporte [00:29:01]:
Is this follow the crypto.org? is that, is that yours?
Molly White [00:29:04]:
Currently it's going to be renamed because it's going to be AI and crypto.
Leo Laporte [00:29:08]:
You know, I think this is where data science is so great. Are you a data science scientist by training, Molly?
Molly White [00:29:14]:
I'm a computer scientist by training. I'm not a data scientist.
Leo Laporte [00:29:17]:
Yeah, but this is what data science can give you, is some real clarity on this stuff. I think this is a really good use of your skills. Good. Well, we'll keep an eye on. So it won't be follow the crypto. It'll be Follow the money.
Molly White [00:29:33]:
I'm renaming it to Tech Influence Watch because it's.
Leo Laporte [00:29:36]:
Oh, I like that. That's good.
Molly White [00:29:37]:
Thank you. You can just follow in the citation needed banner. So I'm sort of centralizing my subscribe
Leo Laporte [00:29:44]:
to Citation Needed support Molly's work and keep up to date on this stuff. It's really great newsletter.
Gary Rivlin [00:29:51]:
Yeah, it sounds great. You could just follow Andreessen Horowitz because they to merge crypto and AI live in the congressional district where Alex Boros, that's who I'm going to end up voting for. Where millions, probably tens of millions are being spent to defeat him. He state legislature. State legislator came up with some pretty mild, I thought, legislation that made sense. And of course crypto AI wanted to defeat it. Did defeat it. And now they're spending all these millions of dollars.
Gary Rivlin [00:30:24]:
He worked for Palantir for a while. He's a data scientist. Alex Burroughs and I get hit mail probably one once a week funded in part by Palantir. Also Andreessen Horowitz OpenAI. Wait, wait. Besmirching him, like how could you vote for this guy? He worked for Palantir and his Palantir literally paying for the. For those mailers.
Molly White [00:30:47]:
Yeah, the crypto industry has done the same thing where they basically are attacking like they ran attack ads against. I think it was Juliana Stratton maybe for trying to. Basically trying to portray her as being funded by ICE contractors.
Leo Laporte [00:31:01]:
They spend $9 million to. To keep her from getting elected.
Molly White [00:31:06]:
Yes, right. But meanwhile like half of the people who are donating to the super PAC that was funding that ad have ICE contracts and it's like who, who signed off on this?
Leo Laporte [00:31:20]:
Let's take a little break. We will have more with this left wing Marxist communist red diaper baby panel in just a little. I don't know, I'm just saying something.
Sam Abuelsamid [00:31:30]:
I've been a lefty all my life
Leo Laporte [00:31:32]:
to keep the red folks sticking around for a little bit longer. Pope Leo, who named himself after he's
Sam Abuelsamid [00:31:44]:
a podcaster in the planet, named himself
Leo Laporte [00:31:47]:
after he's a fellow twit.
Gary Rivlin [00:31:48]:
I do that.
Leo Laporte [00:31:49]:
I wish he hadn't picked Leo because it is a little confusing. But he.
Gary Rivlin [00:31:53]:
Not really to the rest of us.
Leo Laporte [00:31:55]:
No, I think everybody knows the difference.
Sam Abuelsamid [00:31:57]:
Padre probably suggested the name to him.
Leo Laporte [00:32:00]:
Well, it was a little weird. Father Robert was on Intelligent Machines on Wednesday and he kept saying Leo and I said mutt. Oh no. That guy. Okay. Yeah. So he named himself after Leo XIII who was pope at the turn of the 20th century. Between 1800 and 1900 and wrote an encyclical about the industrial age.
Leo Laporte [00:32:18]:
100 years after the beginning of the industrial age, but asking to be more humane in the conversion to industrial life. And I think that's when Pope Leo was elected. He decided to choose that name for that because he wanted to honor that and do the same for our modern times. So we've been kind of waiting for Leo XIV's encyclical about the power of AI. And I don't know, you said, Gary, you had. First of all, I'm not a Catholic. Are any of you Catholics?
Sam Abuelsamid [00:32:54]:
I was raised Catholic, but left at about the age of 13.
Leo Laporte [00:33:01]:
Yeah, my dad didn't want to be confirmed. Yeah, my dad went to Fordham, you know, Jesuit school in high school. But he kind of was a lapsed Catholic too. So my daughter is converted. A lot of people in her age group. She's 34. A lot of people in their late 20s and 30s apparently are turning to the church. Interestingly, it's a trend.
Leo Laporte [00:33:26]:
Molly, are you becoming a Catholic?
Molly White [00:33:28]:
Believe it or not, I have missed that trend. I'm usually so on top of trends.
Leo Laporte [00:33:33]:
Yeah, you missed it. She's actually at a retreat with the Sisters of Providence right now in western Massachusetts, believe it or not. So I, I am myself, and I think it's pretty well known an atheist. I'm not a particularly religious fella. And so I was a little skeptical that the Pope would have anything meaningful to say about AI. Actually, I thought what he said was pretty good.
Gary Rivlin [00:34:00]:
Yeah, I loved what he said. I mean, first off, at least someone in power is talking about this. We need human centered AI, you know, we need, you know, humans in the loop. You know, this. My idea of AI where I can embrace AI is as a co pilot, an amplifier. It's like it's helping me in my work. It's never running things. And I do think there's this thought out there that AI can take over.
Gary Rivlin [00:34:22]:
Take over, you know, as a boss, which is getting a lot of attention, but, you know, take over essential functions. And, you know, that was my favorite part of it, but really my favorite part of it. It was just that he was speaking out about this idea that, you know, we have this small group of tech elites deciding everything. This, you know, Silicon Valley idea, we just got to get a few smart people in the room kind of thing. It just ain't going to work for AI. They themselves talk about how powerful and all, everything this, this stuff is. And so, you know, kind of taking on this idea that, you know, A small group in Silicon Valley in Beijing are going to determine AI for everyone. And by the way, I saw this thought out there in Silicon Valley.
Gary Rivlin [00:35:05]:
Oh, the Pope doesn't understand anything about AI. It's really interesting. Since the mid 2010s, the Vatican has been looking into AI. They sent a group of cardinals, I think it was to Silicon Valley. They met with people like Reid Hoffman, James Monica from Google to start the conversation going, what is, what's its potential? And there's been a group every year going to the Vatican. I know Hoffman goes regularly. Kevin Scott, the CTO of Microsoft, other big names in AI or in tech, goes. So they've really been wrestling with this.
Gary Rivlin [00:35:42]:
I think it is thoughtful document because the Vatican, I mean the Pope and the people helping him with this have been thoughtful about it. I assume he didn't write 42,000 words on his own in so short a period.
Leo Laporte [00:35:54]:
It's so funny because there were people who said, oh, you could tell this is AI written. Which Father Robert assured us it was not AI written. He said, there's no AI involved in this encyclical. This was written by. Yeah, you're right. Not merely Pope Leo, but a committee of church people. And Robert pointed out, and I think this is true, that Pope is not merely a religious leader. He's in many ways a secular leader.
Leo Laporte [00:36:19]:
He's a political figure. He leads more than a billion Catholics worldwide. He is, he has standing to talk about this. And as a humanist, which I think it's safe to say he is, I think it's important. I didn't really disagree with anything he said. He wasn't anti AI, it seemed to me. Right.
Gary Rivlin [00:36:42]:
Well, he was actually very specifically pro AI in the sense that he sees lots of benefits from AI. But you know, I think if I had a boyfriend, boiled down, like the tech isn't the issue where the issue. Yes. You know, like it's, it's like who's controlling this, who's not controlling it? You know, our approach in the last couple years, it breaks in 2023. When I started at the end of 2022, I started looking at AI and kind of through the first half, most of 2023, we were talking as a country about AI, you know, Sam Altman saying, regulate us, the Sanders members of Congress, you know, we're all talking about it and it went poof. And now there's all these things that he brought up, AI and the use of warfare, AI and mass surveillance, you know, AI and the baked in bias. It's trained on Our data, our works. Well, we have bias in our work, sexism, racism, whatever kind of thing.
Gary Rivlin [00:37:37]:
And yet we want to use it for sentencing. We want to use it to decide whether you should get a job, whether you should get this apartment. And also, I don't know, I feel like we started to have a conversation in part, we had a change of administration, but more than that, I think it just became this trillion dollar race to cash in. Google, Microsoft, OpenAI, et cetera, all through safety, over the wayside, and became like, well, we can't do anything to stop this because of China. China, China. And at least that Pope Leo, he gets the story, page one. He's articulating, I think all the right issues will make a difference between, I don't know, but at least somebody's fighting back. At least somebody's articulating kind of a better vision than let's just let this thing ride.
Gary Rivlin [00:38:18]:
And AGI is around the next corner.
Leo Laporte [00:38:21]:
It's, I think, appropriate to step back and take a breath. And that's part of the problem with the way AI has been moving. There's such competition, such a race, as you point out, Gary, competition between the companies, but also between us and China, that nobody's kind of really sitting back and saying, well, what is this? Do we want it? How should we proceed? How should we think about it? What guardrails should we put on it? Molly reminds me of crypto in many ways.
Molly White [00:38:51]:
Yeah, I was just going to say it's kind of the classic idea that any innovation is good. And so you hear all of this talk of we have to protect innovation, which usually means no regulations, but, you know, there's. I feel like no one ever actually decided that all innovation is good. In fact, I think there are plenty of innovations that we would argue are not good. I mean, like, asbestos was not particularly good. Right. And so, you know, there's this idea that if it's innovation, we have to just go with it. And, you know, there's this idea that any technological development is inevitable, and you hear that explicitly from a lot of these people, is that AI is inevitable.
Molly White [00:39:28]:
So just get out of the way. You know, if we don't do it, China's going to do it, so we have to do it. And I think the Pope did a really good job of pushing back on that and saying that, like, ultimately this is a technology developed by humans. We as humans have the ability to decide whether or not to develop it or how to go about developing it. And it's not inevitable. Like, we have the Decision, you know, the capacity to make the decision that this is not something that we think would be useful, or this is not something that we want to hand off to five billionaires who have their own motivations and not much interest in the public good. So I really enjoyed it for that particular reason.
Leo Laporte [00:40:08]:
There is a big backlash against tech going on right now. AI has sped that for about a decade. Well, yeah, but I mean, look at what we were just talking about with data centers, the number of people, the booing at college commencements. AI has become in some ways a whipping boy for all of this technology and all of this big tech. And I think there's a lot of
Molly White [00:40:33]:
helps that I think every tech company is pushing AI, Right.
Leo Laporte [00:40:36]:
So pushing it down our throats.
Molly White [00:40:38]:
Right, right.
Leo Laporte [00:40:39]:
And I. And I am a. I am a proponent. I love AI. I use it like crazy in many ways. I've done a lot of vibe coding. I have an AI agent that knows way too much about me. I'm playing with this because I feel like it's an important technological revolution, but even I acknowledge that there are some big issues and that we really need to be a little bit careful.
Leo Laporte [00:41:06]:
Go ahead.
Gary Rivlin [00:41:06]:
If you look at the polling on this. So Quinnipiac, I think it's called, just did a recent poll, and 65% of Americans are in favor of some guardrails, some regulation. Only 22% agree with this argument. Like, oh, if we put down regulations, we're going to kill innovation. You know, a favorite example is I'm not the only one to use it. You hear it a lot in AI. Those, those guys, you know, there's the Doomers, and I don't put myself on that side. There's the Zoomers.
Gary Rivlin [00:41:36]:
Any regulation is a crime against humanity. Given all the good that AI could do, I. I couldn't. I'd call myself a bloomer. I see potential if we're smart, if we take advantage of our agency here. Because I mean, every foundational technology cuts both ways. I mean, the Internet, the car.
Leo Laporte [00:41:57]:
I mean,
Molly White [00:41:59]:
I use the example, but like it did fireproof.
Gary Rivlin [00:42:02]:
Yeah, well, but we weren't smart about asbestos. Can we be. Can we make it so it's more of a net positive? There will be negative things that happen because of AI. But I still think we're at a point point. We have agency. If we're having discussion, we can make. Make this more of a net positive than a net negative.
Sam Abuelsamid [00:42:18]:
But, Sam, you know, I mean, coming from. From the auto industry perspective, you know, the auto industry has had a huge amount of regulation in the last 60 years. And we've had more innovation during that time, in part because of that regulation than we had in the 60 years before that, the first 60 years of the auto industry. So, you know, I mean, we, our cars are safer, more efficient, more have better performance than they've ever had in the entire history of the industry. And a lot of that is because of the regulation that we put on the industry in. Anytime you're designing something in engineering, having constraints forces you to be more creative. You know, when you're, when you're left unconstrained, you end up with something like the Ferrari Luce, you know, which, you know, we may talk about that later. But the, you know, constraints force, force a degree of creativity and innovation to figure out how can you, how can you work within these constraints and still create a better product?
Leo Laporte [00:43:25]:
And I bet the audio industry, just as the big tech does now, lobbied hard against those regulations.
Sam Abuelsamid [00:43:30]:
Oh, they did, absolutely. It's been to their benefit and it's been to all of our benefit.
Leo Laporte [00:43:36]:
Yeah, yeah.
Gary Rivlin [00:43:38]:
I mean, there's also the trust. There's also the trust factor. I use the example of the railroad where, you know, it was really dangerous when it first came out and the railroad companies opposed any kind of regulation. Government said, no, no, no track size and block signaling and all this kind of stuff. And it made it safe. And by the way, it thrived, not despite regulation, but because of regulation. It just. People trusted the railroad that I'm not
Sam Abuelsamid [00:44:02]:
going to do die same with aviation.
Leo Laporte [00:44:05]:
Aviation, financial markets.
Molly White [00:44:06]:
I mean, like, you hear a lot of, especially in the crypto world, you hear people saying that it's financial regulations that are slowing everybody down. But people trust American financial markets hugely because of the regulations that are in place. And when that regulation goes away, you start to see what we're seeing now, which is declining trust in American financial markets.
Leo Laporte [00:44:25]:
Right. Part of the problem though, with regulating AI is how it's not hard to say, well, a car should be safer and to know kind of how you go about that. I don't even know how you make AI safer. Do we even know what the.
Sam Abuelsamid [00:44:40]:
Clearly, you have to let Donald Trump evaluate all the models before they're released.
Leo Laporte [00:44:44]:
Well, it's interesting.
Sam Abuelsamid [00:44:44]:
Obviously the answer Trump.
Leo Laporte [00:44:47]:
It was part of the way Trump got elected was by promising the crypto Bros and the AI Bros there would be no regulation. One of the first things he did when he got in the White House is overturned. Biden's AI executive order. And yet as soon as Mythos came Out that, that model from anthropic, that anthropic held back because they said if, if everybody got ahold of this, we would have a security cybersecurity nightmare because it's so good at finding cybersecurity flaws. That scared the White House. And I think they, he, he drafted an executive order that said we have to approve all AI models before they're released. And then apparently companies, and particularly David Sachs, his former crypto advisor and AI czar. Yeah, AI and crypto czar, who was put in there basically by Peter Thiel to protect the AI and crypto industries, apparently in an 11th hour call, is, according to, I think the Wall Street Journal, the New York Times to the White House, persuaded him not to put out this executive order.
Leo Laporte [00:45:52]:
But there's definitely this sense in D.C. from Congress, too, that we've got to do something. Question is, what? Yeah, I mean, approving models is not the right answer. You don't want the government to be in charge of saying which models are released and which ones aren't.
Molly White [00:46:10]:
I would push back a little bit on the idea that the promise was no regulation from Trump, because I would say the, the promise was actually, we'll let you write the regulation.
Leo Laporte [00:46:18]:
Okay.
Sam Abuelsamid [00:46:19]:
Captured regulation.
Molly White [00:46:20]:
Yeah. The companies actually really do want regulation that benefits them. Right. And that, that entrenches their position while making it more challenging for competitors to enter the field. And I think that's what we've seen with crypto, and certainly it's what a lot of the AI lobby groups are pushing for as well, is, you know, you have this sort of like competing factions within the AI lobby where there's the open AI and, you know, those folks who are basically all gas, no brakes. We don't want any states to, you know, regulate us and enforce moratoriums on building data centers or, you know, make us liable for harms caused by our products. And then there's the anthropic lobbying group, which is saying, oh, we need regulations that are safety. You know, we need to have controls and guardrails and things like that.
Molly White [00:47:11]:
But both of them are essentially doing the same thing, which is lobbying for their particular interests and sort of building a moat using regulations around their corporate business model, essentially.
Leo Laporte [00:47:24]:
But who has the wisdom to come up with good regulations?
Molly White [00:47:27]:
None of them, that's for sure.
Leo Laporte [00:47:29]:
Or Congress does. Congress. I mean, even Ron Wyden, who's very techno literate. I'm not sure. I don't know who I would say.
Molly White [00:47:37]:
I mean, I think that the thing with congresspeople Is, you know, most congresspeople probably don't have a particularly strong grasp of AI or frankly, most of the issues that they're tasked with legislating. But yeah, they have staff who do. And I have actually found that there are a fair number of Congress people who do have very informed staff. And, you know, it used to be that you would consult, you know, the President's Council of Advisors on Science.
Leo Laporte [00:48:01]:
Well, we had the Office of Technology Assessment too, which Newt Gingrich put out
Molly White [00:48:05]:
of business, and that, you know, that used to be computer scientists and doctors and, you know, professors and all that kind of thing. And now you look at the list of people on that council and there's. I, I was tweeting about it a while ago, or blue skying, skating, whatever you call it.
Leo Laporte [00:48:19]:
Skating. You were skating?
Molly White [00:48:21]:
Yeah, that there are more members of the all in podcast on the Council of Science Advisors than there are professors. It's just all CEOs. So, you know, who, who do you turn to? Right now it's all industry executives.
Leo Laporte [00:48:37]:
I'm not trust professors either, though. I don't know who. I don't know who would know how to.
Gary Rivlin [00:48:42]:
Well, I don't think we should trust any single. Anyone. I mean, I like the idea that we have a dialogue with ethicists and, you know, policy people and, you know, a wide range.
Leo Laporte [00:48:50]:
How about the Pope? Can we trust the Pope? Should the Pope be in charge? He did not make any recommendations, by the way.
Gary Rivlin [00:48:56]:
But, but there are, but, but there are all sorts of, you know, you. For, for these foundational models of a certain size, you need to red team them first before releasing them. You know, I mean, that's an idea.
Leo Laporte [00:49:08]:
You know, that we have lower protections, that no matter what kinds of safety procedures you follow, or guardrails that within minutes Pliny the Liberator will come up with a prompt that bypasses them. It is very hard. I think we've talked about this before on Security now and other shows. The notion of AI safety might be mistaken, that it's not possible to.
Molly White [00:49:35]:
Well, I think that begs the question then, is like, if it's not possible to build a safe model, should we be building the models at all? Right.
Leo Laporte [00:49:41]:
Or should we just acknowledge that they can't be made safe any more than the Internet could be made safe?
Gary Rivlin [00:49:47]:
Well, but then what do we use it for? Do we use it for mass surveillance? Anthropic took a principled stand.
Leo Laporte [00:49:52]:
That would be the smartest thing to do is to say don't use it for things where there is Potential harm because it's not reliable.
Gary Rivlin [00:50:00]:
And there is legislation out there in different states, in Congress, there is legislation, there's kind of intellectual property rules. What can you use for training? How do you compensate those whose material, including mine, including, including probably everyone on this podcast is in there. So there are all sorts of ideas out there. But I think to Molly's point, the power of the crypto lobby, the power of the AI lobby, the power of tech right now means that everything's a non starter. I mean, so, you know, let's give Trump credit. He's like, okay, well maybe we do need to have some kind of regulation out there. I don't, you know, it was never released, so it's hard to critique it. But you just see it wasn't just Sachs, I mean, Musk, some of the CEOs of, I don't know specifically, you know, Google and, or Mark Zuckerberg way Zuckerberg was one of the names that they all kind of like, no, you can't do this.
Gary Rivlin [00:50:54]:
And he pulls it even though it's in front of all of us. Right? This wasn't just some private thing. The announcement was made. And so like, we can't do anything. We can't do anything because industry is so powerful, they have so much money and, and they have the majority of folks in their pocket, so everything will die in the vine.
Molly White [00:51:15]:
Well, and I think it's a really strong indictment of Citizens United because if you look at it from the company's perspective, if they stand to make billions of dollars and the choice is basically, do we spend $100 million on lobbying and super PACs, you know, of course they're going to do it. The payoff is huge. Right. Like it's just simple math.
Leo Laporte [00:51:40]:
Right.
Molly White [00:51:41]:
And so, you know, I think it's really sort of a predictable outcome that we're seeing this kind of thing coming out of the crypto industry, out of the AI industry and certainly I think probably any other industry to follow because, you know, it returns are huge when you can capture regulators, legislators, the president.
Leo Laporte [00:52:00]:
Right. Well, I mean, the, this is a larger rule of American politics that money talks. And when that's the case, you're never going to have a, a society that looks out for the people in society. You only have a society looks out for people with money.
Gary Rivlin [00:52:17]:
Because in the context of this show, it's Andreessen Horowitz that's the number one donor. Not a bank, not the bank industry, not the oil industry. They are the number one giver to the current. Yeah, yeah, yeah, yeah. More than elon during the 2026 election cycle. Yeah, yeah. So. And you know, you know what?
Leo Laporte [00:52:39]:
They looked at Elon and the quarter of a billion he spent to make himself a trillionaire two years later and said that was money well invested and they're doing the same.
Gary Rivlin [00:52:49]:
And they spent 100 million plus.
Molly White [00:52:52]:
I mean, they spent in 2024 as well and had wonderful success with that. So I think it's no surprise that we're seeing them putting money towards again,
Leo Laporte [00:53:00]:
both crypto and AI really, the answer is, and Larry Lessig's been saying this for years to no avail, you got to get the money out of politics before you can even talk about AI regulation or any kind of regulation. Because if. If money's driving policy, it's always going to be policy that supports people with money.
Molly White [00:53:21]:
Right.
Leo Laporte [00:53:21]:
It's just. It's just that's how it is.
Sam Abuelsamid [00:53:24]:
And I think, unfortunately, that's. That's going to be incredibly difficult to do because especially, yeah, we would need to pass a constitutional amendment. And that's because the, you know, this Supreme Court has made it clear.
Leo Laporte [00:53:38]:
Right.
Sam Abuelsamid [00:53:39]:
You know, you know, most of the members of this court, you know, made it pretty.
Leo Laporte [00:53:42]:
They already ruled that they allow Citizen United.
Sam Abuelsamid [00:53:45]:
That. Yeah, that was their statement. I mean, this one third. One third of this court was not on the bench when that was, when that was decided. But, you know, they would clearly support it. Right. So, you know, we settled. We obviously need, you know, some fundamental changes to the Constitution, you know, to get that money out of the system.
Leo Laporte [00:54:05]:
And we've already seen how much courage, how much spine members of Congress have.
Sam Abuelsamid [00:54:10]:
Yeah, well, you know, when it comes to. To do it, I'm not saying it's
Leo Laporte [00:54:13]:
going to happen that make it hard to get elected. They're really, really adamant, I would say,
Molly White [00:54:19]:
is that it's actually hugely popular among
Gary Rivlin [00:54:22]:
left and right, left and right.
Leo Laporte [00:54:24]:
Not just we the people want this.
Molly White [00:54:27]:
Right. And so, I mean, I do actually have some hope in the long term, certainly not the short term, that something like this could happen. Because we see over and over again that when it's put to everyday people, everyone supports it. I mean, Maine had a referendum question in 2024 that passed with 75% of the vote, which is enormous, to basically limit, you know, corporate spending in politics. And that's been repeated in multiple places. And, you know, I think this election cycle, we're seeing more and more candidates who are acknowledging the fact that people are actually pretty sickened by corporate spending in. In politics. And, you See, you know, this wave of candidates who are refusing corporate contributions or super PAC money or whatever it might be.
Molly White [00:55:14]:
So I do think there's something of a shift, and I think that voters are fairly disgusted by any overt spending when they know about it. But I do think that you're right. It's a, It's a big hill to climb.
Gary Rivlin [00:55:27]:
Yeah.
Sam Abuelsamid [00:55:27]:
I mean, the problem is most of it is most of what's being spent is dark money that we don't know where it's coming from, or let's point
Leo Laporte [00:55:34]:
out, what is the money being spent on. The money is being spent on advertising, on brochures. I mean, all money gets you is votes. If the people stop being influenced by what the money's buying and start paying attention to the issues and voting their conscience, the money won't matter. Am I wrong on that?
Sam Abuelsamid [00:55:58]:
Well, you're right in principle. The problem is getting to that point where people do ignore those advertising messages.
Leo Laporte [00:56:07]:
That's a message for our audience because tech people tend to be kind of, ah, your vote doesn't matter. I'm not gonna. I know most. I would bet a large portion of our audience doesn't vote because it doesn't matter. They're cynical. They tend to be cynical. They're smarter, so they tend to be cynical. It's not gonna make any difference.
Leo Laporte [00:56:24]:
My vote doesn't count. So that's what we have to overcome. We have to overcome that sense of apathy, that sense of cynicism, and especially in the people who listen to these shows, because I think it strikes me, these are the people who are particularly cynical about the whole process.
Sam Abuelsamid [00:56:40]:
Yeah. I mean, in 2024, there were as many people who were eligible to vote that did not vote as voted for either party.
Leo Laporte [00:56:49]:
Right. It's always been that way. The huge percentage of people just don't vote. And a lot of them are.
Sam Abuelsamid [00:56:56]:
Well, you know, also, I mean, part of the, part of the problem there is, you know, you know, because of the way our election system is structured, especially at the presidential level with, you know, with the Congress, with the elect, the electoral College, that it's only a handful of states where, where you, you know, where it ever really switches back and forth and where the, where the votes actually make a significant difference in who gets power. Yeah.
Leo Laporte [00:57:30]:
So close.
Molly White [00:57:30]:
But I mean, I would say that, you know, there is significant power in terms of, you know, races below the presidential level.
Leo Laporte [00:57:38]:
Right. Yeah, Right.
Gary Rivlin [00:57:39]:
Well, the problem there is redistricting means that virtually every incumbent is guaranteed re election. They're going to get Reelected by raising all the money because they're incumbents on these important committees. And it is kind of a self fulfilling thing.
Leo Laporte [00:57:56]:
Do your homework, study up. Don't watch those TV ads. Don't follow the brochures. Don't just make your vote.
Molly White [00:58:04]:
Or at least fact check the TV ads. Like that's what keeps me is watching some of the ads from these packs that are like literally just lies.
Leo Laporte [00:58:10]:
Voting is so much work. You know, in California we have, I think, more than 60 candidates for governor in this election that's coming up next Tuesday and a week from Tuesday. And it takes a lot of work to figure out who to vote for. It is nice.
Gary Rivlin [00:58:24]:
And then there's. And then there's ballot initiatives. There's the low.
Leo Laporte [00:58:27]:
You know, it's a ballot. Yeah. And this is just a primary. And people historically don't even vote in primaries, so. And I understand why, because you look at this and goes, I can't.
Sam Abuelsamid [00:58:38]:
It's the tyranny of choice applied to elections.
Leo Laporte [00:58:40]:
Please, we need you to vote and we need you to care and we need you to study and vote your conscience and learn about the issues. Don't just say, well, I saw a TV ad. Because those TV ads are so slanted. And that overcomes money in politics. That's the first step. And then you can start to whittle away at, at the, at the incumbencies and so forth and maybe make a difference.
Gary Rivlin [00:59:06]:
Listen to Leo and Pope Leo both.
Leo Laporte [00:59:09]:
Yeah, Pope Leo and podcaster Leo. Let's take a quick break, do a plug.
Molly White [00:59:14]:
I do have a podcast episode coming out pretty soon with the end Citizens United President who.
Leo Laporte [00:59:19]:
Oh, nice. So where is. Is this a citation needed podcast? Where is it?
Molly White [00:59:24]:
Yeah, it'll be coming out on citation needed.
Leo Laporte [00:59:26]:
Okay, that's great. I would like to, I would like to listen to that citationneeded news. If you go there, there's a podcast feed right at the top and it'll
Molly White [00:59:35]:
come out as well.
Leo Laporte [00:59:37]:
Subscribe. So you get it. Yeah. That is awesome. That is awesome. See, so you're in. You agree with this then this premise. Oh, very good.
Leo Laporte [00:59:48]:
Gotta do something about this.
Molly White [00:59:49]:
Yeah, yeah, yeah. I mean, one of the things that people challenge me on a lot is like, why are you tracking crypto election spending? Every corporation, you know, every industry does it and it's like, yes, they do. That's a problem. And we need to stop at all speaking, not just crypto.
Leo Laporte [01:00:04]:
Yeah, yeah.
Sam Abuelsamid [01:00:05]:
Is it citation needed? The podcast?
Molly White [01:00:08]:
Probably not. It looks citation needed news.
Leo Laporte [01:00:12]:
If you go to the website, another
Sam Abuelsamid [01:00:14]:
podcast with that name, which there's, there's a couple. There we go. Molly White, Citation Needed.
Molly White [01:00:21]:
That's. Yeah, that's it.
Leo Laporte [01:00:22]:
If it's got the name, if Molly White's on the box, then it's the right one.
Sam Abuelsamid [01:00:27]:
CitationNews News. Yeah. Okay.
Molly White [01:00:29]:
Primarily a newsletter and not a podcast. I have basically a podcast version of the newsletter. And so I hope I don't step on the other Citation Needed's toes too much because they are a very good podcast.
Leo Laporte [01:00:39]:
But, well, we, you know, all we care about is, is you, Molly. So listen to Citation Needed DOT News, the podcast with Molly Wood.
Gary Rivlin [01:00:50]:
And by the way, yes, every industry gives money, but what struck me about crypto in 2024 is how much money they gave in how they almost single, single handedly impacted elections. Senate races in, you know, Ohio, what, California if I'm remembering right. You know, I mean, so that to me is like, yeah, every industry gives money, but they're giving tens of millions of dollars there. You know, it's, it is a wave of money that is changing the election completely. As opposed to like, oh, this person has 5 million, that person has 7 million because they have more oil money. You know, crypto to me is like amazing. It didn't, essentially didn't really exist. There's, you know, a bunch of idealists, whatever it was 15, 20 years ago and now they're the one of the most powerful industries in all of D.C.
Gary Rivlin [01:01:40]:
just simply because they are, have all that crypto money right matters.
Leo Laporte [01:01:45]:
You can't get the electorate to do the right thing and, and vote.
Molly White [01:01:49]:
Yeah. But I mean, I think one thing that's really striking is that, you know, everyone knows about Big Oil, everyone knows about Big Pharma. You know, everyone knows that there are corporate influences coming from those industries. But people are very, very unaware of the extent to which crypto is influencing elections. And these days. I mean, people are unaware of the extent to which the President is involved in crypto.
Leo Laporte [01:02:15]:
A crypto has he made in his, in his year and a half in office.
Molly White [01:02:19]:
And just from crypto or altogether? Well, it's over, well over a billion dollars. Yeah.
Leo Laporte [01:02:24]:
In crypto.
Molly White [01:02:25]:
And that's not including. No, that's not including the sort of Trump coin and all price of his crypto holdings because that's all illiquid, just not even worth trying to. Because you'd be in the billions and billions of dollars. But just in terms of like money, like dollars he has cashed out. It's over a billion dollars.
Leo Laporte [01:02:42]:
Yeah. Before the election, the crypto Ball. Yeah, well, actually, that was the inaugural ball.
Molly White [01:02:48]:
It was the inaugural. Yeah. But one thing I was going to say is that, like, there's a crypto media outlet called CoinDesk that just did a poll that found that, you know, the. The majority of respondents were opposed to the idea of an elected official at a high level having personal ties to the crypto industry or, you know, personally profiting from it. But a tiny, tiny percentage were even aware that Trump was so heavily involved. And so it's really just. I think an education issue, in large part is because people just aren't aware of that this is happening.
Sam Abuelsamid [01:03:20]:
Well, I guess I should. That's the other thing that needs to change is that, you know, the fact that we allow politicians, elected politicians, to be making trades on the market. There was a report that came out a week or two ago about how many thousands of trades Trump has made personally in the last year and a half.
Gary Rivlin [01:03:43]:
He's like a day trader. I mean, actually, that was the funniest quote. Like, wow, that's the number of trades of private equity.
Leo Laporte [01:03:49]:
Find the time. I don't know exactly understand.
Gary Rivlin [01:03:51]:
Tweeting it.
Leo Laporte [01:03:52]:
Tweeting times a day.
Gary Rivlin [01:03:54]:
I don't.
Leo Laporte [01:03:54]:
I don't get it. Is he a vampire?
Sam Abuelsamid [01:03:56]:
I mean, just going on.
Leo Laporte [01:03:57]:
Just.
Sam Abuelsamid [01:03:57]:
Was it just last week, you know, they announced this deal, federal government deal with dell for almost $10 billion, and the day before, he bought a bunch of Dell stock.
Leo Laporte [01:04:06]:
Yeah. Right. That's insider trading. Yeah.
Gary Rivlin [01:04:09]:
Yes. Unless you're the President of the United States.
Sam Abuelsamid [01:04:12]:
So, no, no, no. Elected politician. Yeah, no elected politician should be allowed to trade anything during their time in office.
Leo Laporte [01:04:18]:
Clearly. And it's ironic that Nancy Pelosi was so upset by the notion that you might want to curtail her stock trading, but on both sides.
Molly White [01:04:27]:
One thing I think is really funny is that the Senate just passed a change to their rules, basically saying that no senators can get involved in prediction markets.
Leo Laporte [01:04:36]:
Right.
Molly White [01:04:36]:
And it was unanimous.
Leo Laporte [01:04:37]:
They can't do Kalshee, but, oh, we got this. You want to buy some stock.
Molly White [01:04:42]:
That's what I thought was so funny, is, like, there wasn't a single congressperson who would stand up and say, actually, I do want to trade on these prediction markets. But, you know, you try to pass a law that would ban stock trading or crypto trading, and suddenly it's like this hugely controversial thing. But prediction markets, I guess that's where they draw the line.
Leo Laporte [01:04:58]:
We'll talk about prediction markets. There's a lot of news in there, too. All right. I do have to Take a break. Molly White is here. Great to have you. Sam, Abel, Samit, Gary Rivlin. Great panel.
Leo Laporte [01:05:09]:
Lots to talk about. So Google has really doubled down on AI in its search results. And I think you're seeing. Immediately you're seeing a response. For instance, last week, DuckDuckGo installs were up 30%. 30%. And of course, right on the front page of DuckDuckGo, it says, no AI because now when you do a Google search, AI is at the top of the page. AI is everywhere.
Leo Laporte [01:05:41]:
And I think people honestly are nostalgic for the original era of 10 blue links on a page, or at least have the option.
Molly White [01:05:50]:
My mom is using DuckDuckGo, and I did not tell her about it.
Gary Rivlin [01:05:53]:
That's huge.
Molly White [01:05:54]:
Where did you find this? That's awesome.
Leo Laporte [01:05:56]:
Isn't that huge?
Molly White [01:05:57]:
Yeah, it's huge. Yeah.
Gary Rivlin [01:06:00]:
I mean, of course Google is pursuing AI. They should. But what I resent and what I have to assume a lot of people resent is, well, sometimes I just want to Google something.
Leo Laporte [01:06:12]:
Right.
Gary Rivlin [01:06:13]:
You know, and you're just shoving it down my throat. I mean, you know, there's a little tab there was, you know, AI mode. You can just click that. Yeah, I want to do AI. I don't particularly like it. Gemini. I never use Gemini. But the point is, actually, I just misspoke.
Gary Rivlin [01:06:26]:
I sometimes use Gemini because it's right, just right there. I mean, you know, we could mock Google, but they're neck and neck with, you know, monthly users with OpenAI. They're both up around, you know, a billion a month. And it's just because it's right there. You know, my kids, I make fun of them, like, why do you use Gemini? It's like, dad, it's right there. I don't have to download it.
Leo Laporte [01:06:47]:
It's there.
Gary Rivlin [01:06:48]:
It reminds me of, you know, Microsoft taking on that Netscape Navigator, like, well, okay, I guess I could download the Navigator on my Windows machine, but I have Internet Explorer right there.
Sam Abuelsamid [01:06:59]:
And.
Gary Rivlin [01:06:59]:
And that's why they want, I mean, you know, so Meta. Theirs is even more lame. Their AI is. I try to use them all. Theirs is even more lame than Gemini, except they're in the game, too, with hundreds of millions of. Hundreds of millions of users per month. Because it's right there, you know, and I don't want to read.
Leo Laporte [01:07:17]:
They have your information. That's the other side of the equation. They already know Google and Meta both a lot about you. So just as we were talking with this box ad having. It's not just the model, it's having the context to put into the model that makes the model useful. So there's the button right there on every freaking Google property. We use Google sheets for our show rundowns. And it's so annoying.
Leo Laporte [01:07:42]:
Every time I open the sheet, it says, well, you want me to summarize this? Do you want me to eliminate Today it said there's blank boxes. Do you want me to fill those up with something?
Gary Rivlin [01:07:49]:
It's like, no, except for go to Microsoft and Microsoft just has it every
Sam Abuelsamid [01:07:55]:
make up some links and fill them in there.
Leo Laporte [01:07:58]:
Well, Microsoft has, I think, started to see the backlash. They've already announced. We're going to. Oh, yeah, we're going to back off a little bit on forcing copilot down your throat.
Molly White [01:08:06]:
Right?
Leo Laporte [01:08:08]:
So, but Google, see, here's Google's kind of in a rock and a hard place because I think they probably saw some erosion of their search leadership by companies like Perplexity that are doing AI
Molly White [01:08:19]:
searches, which is hilarious because Google has been for decades eroding web traffic to all of the places they're indexing.
Leo Laporte [01:08:28]:
And that's the other side of this. The consequence of this. It's clear Google doesn't want you to go anywhere else. They want you to land on the Google page and stay on the Google page. We'll give you the answer. Don't click that link. You don't need to. We'll tell you everything you need to know.
Leo Laporte [01:08:42]:
Now they're saying for students, we'll draw you the graph, we'll give you the explanation, we'll put graphics in there. You don't need to go to that web page. But it's kind of killing the goose that laid the golden eggs because all the content for those models comes from those pages that they're now disintermediating and putting out of business.
Sam Abuelsamid [01:09:01]:
But it's okay because they're going to make up new pages. They're just going to generate new stuff to fill.
Leo Laporte [01:09:07]:
Oh, you've got a blank in your web. We'll just fill that in.
Molly White [01:09:11]:
I mean, it's already happening, right? Because they're scraping so much of the web and so much of the web now is AI generated that you're getting this sort of ouroboros of AI generated content that's being trained on AI. And there were some really interesting papers about model collapse when, you know, you basically feed an AI too much AI content as training data and things just go haywire.
Gary Rivlin [01:09:33]:
Mustafa Suleiman, the head of AI consumer AI for Microsoft, said they're not going to use, you know, to the extent they can. As you're pointing out, who knows, knows, you know, what's AI slop and what's human slop. But it's, they're going to try not to use the fabricated data, the AI data, because it's so unreliable. But you know, it's like from the corporation's point of view, like every company is doing this. Every large enterprise software person there, you know, a company, they're, they're putting, if we give you your AI agent, it's right here, it could do this, it could do that. You know, it's kind of the, this, this new UI that instead of having to go anywhere, you just, wherever you are, there's all the tools you want. I mean, this seems one of these things that's really hard to stop. It drives me crazy.
Gary Rivlin [01:10:20]:
There might be a little bit of a backlash, but I bet if we did this show a year from now, it's going to be more integrated, not less.
Leo Laporte [01:10:28]:
Yeah, yeah. Because in the long run, even if there's consumer backlash, people are going to still use Google to search, aren't they?
Sam Abuelsamid [01:10:38]:
I don't know. I mean, based on what they showed, you know, where now you just get a page, you know, with a window that grows as you fill in your question, and then it'll just give you an answer. I don't know that. I don't know that I want to continue using Google. Do you still use Cogi?
Leo Laporte [01:10:56]:
Yeah, yeah. I pay 25 bucks a month to use an alternative to Google that just gives you links. Now Kagi actually has a really good perplexity style AI orchestration engine too. But I don't use that so much. I just want Google without the ads with it. I just want the links. Just the links, man. But also, I'll be honest, increasingly I don't go to a search page, I don't type in the browser window, I don't type a query.
Leo Laporte [01:11:23]:
I use my AI agents to do all of that stuff. And my AI agents have MCP servers that, that query the web. You know, I'll do, I'll say things like, give me the top 10 AI stories. This, you know, in the last 48 hours. And it goes out and looks at X and Reddit and hacker news and, and my RSS feeds and comes up with the top stories, the stories that we're reading today. Almost all those links were AI generated through the week. I don't even, I gave it my RSS feeds. So I feel like I've noticed.
Leo Laporte [01:11:58]:
I'm Doing a lot less searching.
Gary Rivlin [01:12:01]:
I mean, Google and Facebook help, you know, damage seriously, you know, news sites because people weren't going to them as much and now it's, they're going to go to them even less. There are some statistics that, you know, like, you know, there's 58% lower click through rates with AI. And you know, I'm guilty of that. I mean, you know, sometimes I want to check it, I'm writing about something, I need to check it. But you know, what happens to these sites? Who are the content creators if increasingly, with each increasing year they're going to have less traffic, that is less money. It's just, I don't know, I just kind of play it out in my head.
Leo Laporte [01:12:36]:
I feel a little guilty because we rely on all of these sites. I try to mention, when I'm quoting a site, I try to mention the site, often try to mention the author. But honestly, I'm not doing a lot of enterprise journalism here. I, I just go out and scrape sites and these stories come from them.
Gary Rivlin [01:12:56]:
I think that's why AI wins. It's just so easy, it's so convenient. We're all in a hurry, we're all overwhelmed.
Leo Laporte [01:13:03]:
The difference is this is a human curated experience. Admittedly, I'm using AI, but then I go through and say we're going to talk about this and this. I feel like human curation still has some value, but maybe not for much longer.
Molly White [01:13:15]:
I don't know.
Gary Rivlin [01:13:16]:
Yeah, I actually think, I actually think that is the secret weapon.
Leo Laporte [01:13:20]:
It's, it's what all of us do. That's what all of us on this panel do. Right? That's what you do, Gary, when you write your book. That's what Molly does when she writes citation needed. That's what Sam does when he does his analyst reporter does his wheel bearings podcast. What we do is people follow us because of our curation, our taste, our instincts, all that.
Molly White [01:13:41]:
It's something that the AI companies underestimate a little bit, is that people really do like human created material, whether it's, you know, music or artwork or a newsletter or, you know, the news. People do value that. And you know, I think that's why you see some of the backlash. But you also see people, I think, who seek out actual human made material as well. And I think that that's, that's going to continue to be the case even as AI content proliferates.
Leo Laporte [01:14:13]:
Do you feel like maybe I should. I mean, I, I use AI, so I, I mean, I, I'm, I'm An AI fan. I'll make that clear. I. But I use it as a. As a tool to do this human stuff that we do. I would never create a podcast that just was, you know, AI voices with AI panels.
Gary Rivlin [01:14:36]:
Those exist.
Leo Laporte [01:14:37]:
Oh, they totally do. Look at NotebookLM. I mean, they exist, and I think there'll probably be more and more of them, to be honest. Spotify's doing it. Audible. Does it. I imagine there'll be more and more of them.
Gary Rivlin [01:14:48]:
But I've watched several podcasts of mice of. You know, here's a podcast about Gary Ribland's new book, AI Valley, that I learned because I have, you know, the Google ego search. You know, I got a little a message like, oh, wow, what did I
Leo Laporte [01:15:02]:
have to say about that topic? Are they.
Gary Rivlin [01:15:04]:
Yes, it's all a generator and, you know, and kind of in defense of humanity. It's like, I mean, they're. I'm sort of impressed that it could do anything like that.
Leo Laporte [01:15:13]:
But, like, that's more my experience. Like, I'm not saying it's great. It's just amazing that it can do
Gary Rivlin [01:15:18]:
even that, but it was really lousy. It was, like, so quirky. Like, it just. Why'd you choose that? That's not a particularly interesting. But I. I mean, I was saying this. I was saying this before when we're talking about the Pope, like, you know, if the human's at the center, using AI. Use AI all day, every day for my work.
Gary Rivlin [01:15:38]:
But it's me, you know, I never have it. You know, write me this article, read these interviews, write the article. It's more as I'm writing the article, I'm really struggling with this paragraph. There's something wrong with it. Tell me what's wrong with it kind of thing. It's just really interesting. But it's me. It's not like someone else can write that same article using the same engine.
Gary Rivlin [01:15:56]:
I do. And so to me, if it's, you know, human centered and it's you as a curator, your judgment, your ethics, your instincts, whatever those things are, given the task, you know, I'm cool. I'm cool with that. Actually, though, it's interesting. I was having a friend writing for the New York Times. Now, if you're freelancer, you have to sign something and say, I will not use AI at all in the creation of this article. Which, you know, I. I understand research.
Gary Rivlin [01:16:26]:
You know, I didn't read it myself. I asked that same question. He said, I don't do it because I'm scared that I'm Breaking the rule. But it's just like, you know, I mean, I get where they're coming from.
Leo Laporte [01:16:37]:
They don't want to typed every word on the page. I don't think that it would be a bad thing to use AI for, to inform yourself.
Molly White [01:16:46]:
Well, it depends if you fact check, right?
Leo Laporte [01:16:49]:
Well, yeah, that's true. You gotta fact check. Do you think it is our obligation, the four of us, to put more human created content into the content sphere, into the Internet? Like as human beings, we gotta fight the fight and make human created content and put it out there?
Molly White [01:17:10]:
I don't think it's my obligation, but
Leo Laporte [01:17:12]:
it's something I should be. It's your duty. It's your duty as a being.
Gary Rivlin [01:17:17]:
It's a hard question to answer because that's simply what we do.
Leo Laporte [01:17:19]:
It's what we do anyway kind of thing. By the way, that's the other point. That's the point Benito, our producer, is always making. He's a musician. So people make music because they like to make music. Writers write because they like to write. You're not gonna stop people from creating.
Sam Abuelsamid [01:17:34]:
Right.
Gary Rivlin [01:17:35]:
I think the new world is gonna have, you know, there's AI created art and there'll be some people who really like it. Some of it probably going to be really interesting, but I think people are going to prize human centered art much more because of it. Just like, you know, when digital music came along, you know, suddenly, like records
Leo Laporte [01:17:52]:
or this bespoke thing, kids are on vinyl heads.
Molly White [01:17:57]:
Well, I mean, you do like.
Leo Laporte [01:17:59]:
Yeah, I knew it. I somehow I knew it.
Molly White [01:18:00]:
Yes, but like teenagers are walking around with disc men.
Leo Laporte [01:18:04]:
Yes.
Molly White [01:18:05]:
Like CD players now. And you. There is this really strong backlash, I think, to not just AI, but like tech in general. You know, they don't want to be on their phones all the time. They don't want to be connected to the Internet all the time. They want the more analog lifestyle, which I can respect.
Gary Rivlin [01:18:23]:
I think what's really interesting, the people, the people who are harshest about AI aren't older folks, which would have been my guess. It's the college kids who are bullied.
Leo Laporte [01:18:33]:
They are, you know, it's like this
Gary Rivlin [01:18:35]:
is coming right for them.
Sam Abuelsamid [01:18:37]:
Yeah. They're worried that they've got their whole life ahead of them.
Leo Laporte [01:18:39]:
Yeah.
Sam Abuelsamid [01:18:40]:
That's going to be dominated by this stuff. I mean, you know, at least I
Leo Laporte [01:18:44]:
would say to them, don't worry.
Sam Abuelsamid [01:18:45]:
A couple of us are, you know, closer to the end than the beginning of our careers.
Gary Rivlin [01:18:51]:
No comment. No comment.
Sam Abuelsamid [01:18:53]:
So I have a friend I know how Old I am. Hi, this is Benito.
Gary Rivlin [01:18:57]:
So I have a friend who has a kid who's I think seven or
Sam Abuelsamid [01:19:00]:
eight, and this kid loves Suno.
Gary Rivlin [01:19:03]:
So I think there is a generation that are like really young that love
Leo Laporte [01:19:07]:
to prompt their music.
Molly White [01:19:08]:
Yeah, a gap generation.
Gary Rivlin [01:19:09]:
There's going to be a gap generation.
Leo Laporte [01:19:10]:
Suno's pretty good. I enjoy Suno. There are people, we were talking to Harper Reed. There are people who only listen to music created by AI.
Gary Rivlin [01:19:20]:
But I think those people also don't
Sam Abuelsamid [01:19:22]:
give human music the same chance that
Leo Laporte [01:19:25]:
they give AI music, maybe, or it's just a lot of work.
Gary Rivlin [01:19:29]:
No, they won't sit and listen to an album from a human, but they'll listen to 10 AI tracks.
Sam Abuelsamid [01:19:34]:
They'll sit and listen to 10 AI tracks.
Gary Rivlin [01:19:36]:
How is that less work?
Leo Laporte [01:19:37]:
It's more complex, it's less anodyne, it's less homogenized. It's. It's. There's too much.
Sam Abuelsamid [01:19:42]:
You don't have to think about it.
Leo Laporte [01:19:43]:
Yeah.
Sam Abuelsamid [01:19:45]:
You know, in the camera, you know, average of it's like preaching collective.
Leo Laporte [01:19:51]:
I don't want to chew, I just feed me Soylent.
Sam Abuelsamid [01:19:56]:
Exactly. Yeah.
Gary Rivlin [01:19:59]:
When the camera came along, it was like, oh, no, it's going to destroy art. Because why do you have to paint the landscape? You just take a picture of it. And I think what's interesting is now photography is an art form. You know, art is still thriving. And I assume that the same thing is going to happen with AI. Like, people will love AI music, I guess, but there'll be people who seek out human. It sounds different, the imperfections, you know, I'll throw my 17 year old, then 15 under the bus, like, you know. Did you write that paper about, you know, Lord of the Flies? Yeah.
Gary Rivlin [01:20:33]:
Read it, dad. And within three sentences I just like. Which model did you use? The color drains from his face because like, it was too perfect. It was just like. It was just obvious. It's like it's either, as Sam was saying, it's just lowest combinator flat is this thing, or it's just like no human rights or talks like that.
Leo Laporte [01:20:50]:
But how long. That's the uncanny valley in prose, essentially. How long before it gets good enough that you don't. There's no uncanny valley.
Gary Rivlin [01:21:01]:
We're still there with CG and movies, though. We're still there in CG with movies.
Leo Laporte [01:21:04]:
We are. We're still, I think sooner music. A lot of people can't distinguish Suno music from real music.
Sam Abuelsamid [01:21:09]:
Well, and, and to be fair, you know, probably most of what's Written by humans would fall into that, you know, lowest common denominator, you know, uncanny valley.
Leo Laporte [01:21:19]:
Yeah.
Gary Rivlin [01:21:20]:
There's a lot of human slop. A lot of human slop out there.
Leo Laporte [01:21:22]:
That's great. Slop. Better than anybody.
Molly White [01:21:24]:
I mean, there's a reason the AI sounds like that, right?
Leo Laporte [01:21:27]:
That's right, exactly.
Sam Abuelsamid [01:21:28]:
It's just averaging all of that.
Molly White [01:21:29]:
Yeah.
Sam Abuelsamid [01:21:30]:
But, you know, there, there are exceptional writers out there, like Gary, like Molly, you know, that. That do really special, you know, that do something really distinct and different.
Leo Laporte [01:21:40]:
I think that as time goes by, we will. They'll be an equilibrium. I think we're going to see some pendulums swinging back and forth, anti AI, pro AI. And then there'll be some equilibrium. You know, I've just started watching because they were talking about it on Wednesday on the show. Have you seen the show Humans? It's on Amazon Prime. It's three seasons, started on BBC4.
Sam Abuelsamid [01:21:58]:
Is this the one about the robots that it's.
Leo Laporte [01:22:00]:
Yeah. So the premise is it's near future and that we've come, We've come. We've actually been able to make bipedal humanoid robots and they mostly are domestic servants, although there are some sex bots and there are bots doing laboring in the fields and so forth, and they're doing the jobs that humans don't want to do or that are very hard or difficult. It's good because it isn't 100% anti bot. It's not pro bot either. And it raises a lot of the issues that will come up, I think, in the near future over all of this. You know, it starts. I've just started watching.
Leo Laporte [01:22:37]:
It starts with the husband. His wife is away for work and he's got three kids and he's trying to take care of them. And it's a lot of work. He's got his own job, so he buys one of these bots because somebody's got to clean and cook. And the wife comes home and says, what, are you replacing me? He says, no, no, I just want more time with you. I think philosophically it's very interesting and as somebody who is very much a pro AI user, it has kind of opened my eyes a little bit as to some of the difficulties that this is going to.
Gary Rivlin [01:23:08]:
So, Leo, I want to get back to the question you sort of posed, like, will GPT 10, will it be able to write a beautiful novel the way a human can?
Leo Laporte [01:23:18]:
What do you think?
Gary Rivlin [01:23:18]:
You know, I mean, first off, I think we're hitting some limits. We've, you know, These models have ingested all of human creation. There's, there's no, there's no more stuff right now. It's AI slop. This whole AGI thing, I think we're a breakthrough or two away before we get what Silicon Valley is saying we're going to get. But I think it's going to keep on pac manning up. If right now, I would say AI if you write a basic report, if you're writing a press release, if you're doing a 400 word hot take your toast. I'm sorry, these things are good at doing that.
Gary Rivlin [01:23:55]:
Can it write a basic news article? Yes. But can it write a feature? No, they're terrible. I've tried, I've played around with it. It'll get better at it. But I am convinced that GPT 10, whatever, just, you know, some point 10 plus 20 years into the future, I don't think it could ever do it. I think that's the missing ingredient. You know, AI is obviously, you know, artificial intelligence, but I think it's alien. Alien intelligence.
Gary Rivlin [01:24:20]:
It knows everything and understands nothing. It, it has no common sense. A five year old understands so much of the world that these things don't.
Leo Laporte [01:24:28]:
So you don't think we'll ever bridge that uncanny valley?
Gary Rivlin [01:24:32]:
You know, I'll be dead by the time we do it. If you ask. You know me back to Sam's point where older journalists here. But it's. I'll let it go, Sam.
Sam Abuelsamid [01:24:42]:
No, I was including myself in that group.
Gary Rivlin [01:24:46]:
But you know, I guess I am. Dude, there's a really fun thing the New York Times did a year ago.
Leo Laporte [01:24:52]:
Oh, I know what you're going to talk about. Yes,
Gary Rivlin [01:24:57]:
Curtis Sitfield. I mispronouncing your name. My apologies. But they had this really accomplished novelist, short story writer, go up against a machine and you know, I'll tell you the short story with the same prompts. They all had the same prompts. That would be, you know, a summer romance with flip flops, blah blah, blah kind of thing. And you know, I happen to have read the AI one first. I was like, wow, this.
Gary Rivlin [01:25:18]:
Yeah, this is good. Is this her? And then the moment I started her story like within three or four sentences, like, oh no, this is the novelist, this is the real writer. It's that, you know, by definition it's flat, by definition as low as common denominator.
Leo Laporte [01:25:33]:
But a couple of months ago, I thought you were going to talk about this. The New York Times, Kevin Roose and Stuart Thompson did a quiz. Who's a better writer AI are humans, and they have passages, and they asked people. And more than half of the people, regular people who read them, like, preferred the AI version and couldn't distinguish between the one or the other. I actually did it. And the only reason I got almost all right, I missed one was because I had read some of the books.
Molly White [01:26:06]:
I was gonna say if you. A lot of them, you can recognize.
Leo Laporte [01:26:08]:
I recognize the book.
Molly White [01:26:09]:
Yeah.
Leo Laporte [01:26:11]:
Which helped. But the AI writing was pretty good. Right, Molly? I mean, did you do this?
Molly White [01:26:17]:
I did it. I was. I had the same problem that you did, which is that I recognized a lot of the books, but I also. Definitely. A lot of it read AI to me.
Leo Laporte [01:26:27]:
Did it. Okay.
Molly White [01:26:28]:
Yeah.
Leo Laporte [01:26:29]:
I may. I may be tone. I may be tone deaf when it comes to AI. I may actually not have the detection abilities you guys have, and it's how
Molly White [01:26:38]:
much that varies between people. Also, like, I feel like some people have a really good spidey sense for it, and some people cannot tell whatsoever.
Leo Laporte [01:26:46]:
I like wordplay, and a lot of the AI stuff had kind of bizarro wordplay, which maybe have won me over a little bit. Like, they often do weird similes and things, and I kind of enjoy them. Right.
Gary Rivlin [01:27:01]:
I think there might be an age element. I mean, I'm using a, you know, universe of one or a pair here. My son and I, when this is 2023, there was some site and you had a minute to decide, are you talking to a human? Are you talking to a machine? I'll confess, I was kind of 50. 50. I couldn't tell. But he was, I guess, 12 at the time, and, like, he was getting, like, 80, 90%. Right.
Leo Laporte [01:27:29]:
Well, that's encouraging. That's.
Gary Rivlin [01:27:30]:
And so, like, I mean, I don't know. I just. I just kind of assume from that, that we're talking about, you know, that a young person's better, they're more native, whatever. But, you know, I mean, I don't know. I haven't seen any studies on that.
Molly White [01:27:44]:
Yeah, I'd be curious. Study on it. Like, how much of it is age related versus, you know, level of technological.
Leo Laporte [01:27:51]:
Well, I'm older than all of you, and I'm easily fooled, so I guess that that says something. I don't know. Yeah. Maybe if you grew up, if you're AI native, you can tell the difference. I don't know. Time for another break. Molly White is here. Citationneeded News.
Leo Laporte [01:28:10]:
Subscribe now. Listen to the podcast coming up about Citizens united. She's also mollywhite.net and you will see the announcement when she updates the bitcoin giving this cryptocurrency giving to all, technology giving. I think that'll be a really good site. I'll make sure we plug that. I really appreciate what you do with those websites. Mollywhite.net Gary Rivlin, also here. Pulitzer Prize winning, not novelist, nonfiction writer.
Leo Laporte [01:28:37]:
Have you written a novel? I mean, you've written many books.
Gary Rivlin [01:28:40]:
I have never written a novel. My 17 year old asked me, like, dad, when are you gonna write something interesting that you know, like fiction?
Leo Laporte [01:28:46]:
So I love your stuff. The book you did on Katrina was. That's the one you won the Pulitzer for. Incredible. The book Broke usa about debt, was just amazing. Thank you. I like your nonfiction, but I feel like, don't they say that everybody's got a novel in their desk drawer next to.
Gary Rivlin [01:29:07]:
I have neither, but I like bourbon to get no work done. But actually, this summer, it's funny you should say this. This summer he decides he wants to be a screenwriter. God help me.
Leo Laporte [01:29:18]:
Oh, good.
Gary Rivlin [01:29:19]:
And this summer, we're gonna write together. I mean, right in the same place together. And I'm thinking, like, maybe, maybe this will be the moment when I do a fiction, a work of fiction. But to now, not then.
Leo Laporte [01:29:32]:
You wouldn't be kind of competing with him in your own turf. You'd be kind of on another, you know, unfamiliar ground. It would probably be a good idea. I like it.
Gary Rivlin [01:29:40]:
Given he's 17. He's like, well, dad, I'm better at storytelling than you. I mean, he literally said that to me this week, and I laughed at
Leo Laporte [01:29:48]:
him, God bless him.
Gary Rivlin [01:29:50]:
And he stuck by it.
Leo Laporte [01:29:51]:
You know what we did on Friday night? So there was 4chan. Had a 4chan.
Gary Rivlin [01:29:58]:
Yeah, he's been here Friday night.
Leo Laporte [01:29:59]:
Bear with me on this.
Sam Abuelsamid [01:30:00]:
You're not Pope Leo.
Gary Rivlin [01:30:01]:
You're definitely not Leo.
Leo Laporte [01:30:02]:
It's starting a little weird. Okay, I understand they call it a creepypasta. You know, copy pasta, where you paste something. It's a creepypasta. Some years ago, they had a creepypasta of a kind of an empty room, yellow walls, kind of this liminal space. And then there was this guy named Kane Pixels who turned it into a viral video called Back Rooms. And there were a number of YouTube viral videos he created using Blender and others about these kind of weird spaces. Well, Hollywood said, you know, we need more kids coming to movies.
Leo Laporte [01:30:37]:
So this guy is 20 years old. His motion picture Back Rooms debuted on Friday. We went to the opening because he's from our little town here. He Was there my 23 year old's a big fan. Got his picture taken with him. And this is now the fourth fastest highest grossing first weekend for a horror movie ever. They made it for $10 million. It's grossed more than 30 million already in three days.
Leo Laporte [01:31:04]:
It's called Backrooms. It's 20 year old director. So Gary, you're a teenager in three years, maybe writing that novel, still moving. Might have that Booker prize in hand. So I don't know.
Sam Abuelsamid [01:31:18]:
Okay.
Leo Laporte [01:31:18]:
I don't know. You know, it's just, it happens. We're in a, you know, this is, this happened. So you're old enough, Gary. I certainly grew up in this because I was a baby boomer. It was a youth culture when I was growing up in the 60s and 70s, right, the youth dominated everything. And then we continued to dominate it right up to the present. And now it's a bunch of 70 and 80 year olds dominating, dominating the world.
Sam Abuelsamid [01:31:44]:
I wonder. Grip on power. You never want to let go.
Leo Laporte [01:31:46]:
You don't want to let go. We boomers, we love it. I'm wondering though, we're gonna all die soon. Maybe the swing back to youth is starting to happen. I think this might be people like
Molly White [01:31:57]:
me who are stuck in the middle are just gonna always.
Leo Laporte [01:31:59]:
I feel bad for you, Molly. I do.
Gary Rivlin [01:32:01]:
Sorry.
Leo Laporte [01:32:01]:
I'm so sorry, Molly. We ruined it for you. We boomers, really ruined it for everybody. And I apologize on behalf of, of my cohort. We're going to take a break, come back with more Samoa. Samit is also here. I, we. I want to ask you about the luche Johnny Ives Ferrari, which apparently is getting some bad reviews.
Leo Laporte [01:32:23]:
Have you driven it?
Sam Abuelsamid [01:32:25]:
No, nobody's, nobody outside of Ferrari has driven it yet. Although probably some of their best customers have had a chance to sample it.
Leo Laporte [01:32:31]:
But how much is it?
Sam Abuelsamid [01:32:33]:
$640,000.
Leo Laporte [01:32:34]:
Okay, well, it won't be me driving it, that's for sure.
Sam Abuelsamid [01:32:37]:
But I do have a novel in my drawer, but I didn't write it.
Leo Laporte [01:32:43]:
Do you have bourbon in your drawer? I was visiting a friend named Jeff, Jeff Atwood the other day. And on his kitchen counter he had what looked a lot like Infinite Jest. You know the book, the long book everybody pretends to have read. And I said, oh, who's reading Infinite Jest? And they said, no, no, no, look closely. It wasn't Infinite Jest, it was Infinite Jeffs. Somebody as a joke, and you can buy it on Amazon, by the way, has made a book that looks just like Infinite Jest. But the text of it, the Blurbs, the front cover, the back cover. All it is, is Jeff, Jeff, Jeff, Jeff, Jeff, Jeff.
Molly White [01:33:25]:
AI could never do this, right.
Leo Laporte [01:33:27]:
This is brilliant. You can buy it for 25 bucks on Amazon. If you know somebody named Jeff, this would make a great gift. That's all I can say.
Gary Rivlin [01:33:37]:
Are you saying if you open it up, the only thing you see is the word Jeff repeated?
Leo Laporte [01:33:41]:
Yes. You want to read a sample here? Let's just. Let's just open it up.
Gary Rivlin [01:33:46]:
Don't give anything away. Don't give anything away.
Leo Laporte [01:33:49]:
Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff. The dedication. Jeff. The first chapter. Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff, Jeff. It is. It is like 800 pages of the word Jeff. And by the way, I don't even
Molly White [01:34:06]:
know anyone named Jeff, but I kind of want to copy them.
Leo Laporte [01:34:08]:
You want to know somebody named Jeff now, by the way, it says on the. On the Amazon, excerpt copyrighted material. So don't steal this.
Gary Rivlin [01:34:17]:
Written by. Written by.
Leo Laporte [01:34:18]:
If somebody wants to write, you know, Molly. I don't know, Molly Jones diary. I don't know. You could do that. The back cover is great. Here's the blurbs. They're all by Jeff, and apparently it was written by a cat named Jeff. So now you don't have to buy it because you've read it.
Leo Laporte [01:34:41]:
I hope AI ingest this. I'd like to see what AI does with this. All right, that's the novel in my drawer. And no, there's no bourbon next to that either, unfortunately. You're watching this week in Tech, Sam Abel. Sammit. Gary Rivlin, Molly White. Sam.
Leo Laporte [01:34:57]:
A week from Monday, it's wwdc. We will be covering that conference, and we expect, speaking of AI, Apple to clarify, shall we say, its position on AI on the iPhone. They will also open show iOS 27 Mark Bloomberg. I'm sorry, Mark Gurman at Bloomberg, although I think of him as Mark Bloomberg got a leak of the look of the new iOS 27 series. Going to be in that little window up at the top, living there all the time. These are illustrations, not actual photos, but illustrations created by Bloomberg showing the revamped Siri interface, a new Chatbot style app, and other major changes that Apple will announce a week from Monday. Gurman writes the images are based on information viewed by Bloomberg and people with knowledge of the company's plans who asked not to be identified because the software isn't yet public. And of course, Apple declined to comment.
Leo Laporte [01:36:03]:
But I think one thing that we know for sure is that AI will take front and center at, at the keynote at WWDC and on the new iPhones. And suddenly for a lot of people, maybe your mom, Molly, who aren't really used to using AI, it will be on their phone. It will be a big presence. And I think that is going to be a sea change, I expect, in how people feel about AI and maybe help its acceptance a little bit. What do you, Molly, you think your mom will. Does she use an iPhone?
Molly White [01:36:38]:
I believe so, yes. But I mean, I would say that AI is already kind of ubiquitous.
Gary Rivlin [01:36:43]:
Right?
Molly White [01:36:43]:
Like if you use Gmail, it's trying to summarize your emails. If you search something that's coming up, you know, I don't know how much it will change to have one more, you know, insertion of AI into your everyday life.
Leo Laporte [01:36:55]:
But the only reason I'd say it might be different is because Apple is notorious for being good at productizing this stuff in a way that is not as offensive. Look at what Google and Microsoft have done has become offensive to people. That's why your mom's using DuckDuckGo. Yeah, I think Apple maybe has a lighter touch. I don't know, maybe not. Maybe they, I mean, backlash.
Molly White [01:37:16]:
I feel like people were really annoyed by the attempts to summarize your text message or whatever it was, you know, that was just always gone very well.
Leo Laporte [01:37:26]:
Yeah, yeah, exactly.
Molly White [01:37:27]:
Right.
Leo Laporte [01:37:27]:
They were just bad and I left them on because they were hysterical. They still are hysterical.
Molly White [01:37:32]:
Yeah. Maybe some suspicion.
Leo Laporte [01:37:35]:
I see this all the time. I just got it again. There was motion at your front door multiple times. Yeah.
Molly White [01:37:43]:
Is it your cat?
Leo Laporte [01:37:44]:
Yeah, probably. But it's like that's the summary. And if the first time you see it it's like what? And then you realize, oh no, it's just wrapping up about a bunch of kitty motion. Wow, there are quite a few. Okay, maybe Apple wasn't so wrong on that one really wants to come inside that cat. That cat really wants to come in. Holy moly. Okay, somebody's gonna have to answer.
Molly White [01:38:08]:
Dancing in front of your friend.
Leo Laporte [01:38:09]:
I don't know what's going on. All right, you know what? Apple got it right. There were multiple. I'm scrolling through people downplayed it a little bit.
Gary Rivlin [01:38:20]:
So I mean, to your non cat point here, Leo, I do think that these huge companies with billion plus platforms or platforms with a billion plus people on it. We talked about meta before, we talked about Google before and I think Apple falls in the same category. Yes, we might resent it. Yes. We're a bunch of tech reporters, so we in particular are resenting it. But I just have to think if it's there in front, people are going to use it. And I think Apple, you know, the story on Apple is they've just been fumbling it every bit of the way. They were absent in 2023, 24, 25.
Gary Rivlin [01:39:01]:
Now in 2026, they're yet again going to unveil their AI strategy. And you know, I'm dubious of that. I'm sure they're going to be using other models.
Leo Laporte [01:39:10]:
Models.
Gary Rivlin [01:39:10]:
And they're going to use Gemini.
Leo Laporte [01:39:11]:
They're going to use Google's model.
Gary Rivlin [01:39:13]:
Well, they're, Google's paying for them to use Gemini. And just you, you can opt for others. You can, you know, if you choose, you can have chat, GPT, whatever underneath it. But my point is by just having in a box right there, I, I do think it becomes more ubiquitous in the sense that more and more people are using it. Yes, Molly, AI is everywhere we talk about. We can't help but think about it. But that doesn't mean, you know, everyone's using it. I just think Apple is another big step in a larger portion of the planet using AI because it's right there.
Sam Abuelsamid [01:39:45]:
Yep, that's kind of, kind of hard to avoid.
Leo Laporte [01:39:50]:
Have they given up on Siri? Because if people still use Siri, Siri is going to presumably suddenly be smart and useful.
Gary Rivlin [01:39:59]:
Yeah, Alexa and Siri, you know, it's funny, like, so basically those, you know, Amazon and Apple were way ahead of most everyone on AI. Right. I mean, you know, that's what that is. And I just assume like, oh, okay, well, you know, it's more rules based. It's not, you know, machine learning. But they'll make the shift and it's a lot harder to do than you would think to give it intelligence. I mean, you might as well like, you know, we've seen this with technology a lot that, you know, kind of of bolting something on something else is never as good as something that's built from scratch. And I just think that's Siri's problem and that's Alexa's problem.
Gary Rivlin [01:40:37]:
Supposedly Amazon has taken that on and finally they have a better product. I don't use either of those, so I have no idea. But it's taken a lot longer than I would have thought to make the jump from kind of the old fashioned. Okay, here are the responses to this question as opposed to the conversational aspect that we're all used to now with chat.
Sam Abuelsamid [01:41:00]:
But, well, and you know, I'm an Android user and Gemini is far less functional now than Google Assistant was before. There are so many things I, you know, the things I used to use Google Assistant for Gemini does not work like, for example, you know, I, you know, I have garage door opener that is, was, you know, was compatible with Google. I used to be able to open that, you know, from the car with, you know, using Google Assistant. That does not work with Gemini. I have, we have some LED lights around that notable failure for Google. Yeah, Automations that do not work with Assistant. Yeah, yeah. And so there, there's a bunch of functionality that just isn't there, you know, and that's, this is one of the things that really annoys me about Google is that they will, they'll take stuff that was working reasonably well and they say, no, we're going to replace this with something else and you lose half the functionality or more from that thing.
Sam Abuelsamid [01:42:02]:
And it takes them years to get it back to where it was. You know, they did this with Google Play Music when they went to YouTube music. And you know, they did it with Assistant going to Gemini. And they've done it with lots of things over the years and it's been very, very annoying,
Gary Rivlin [01:42:17]:
I think.
Leo Laporte [01:42:18]:
So as somebody who's spent a lot of time building an agent, you know, this was going to be the year of the agent started with Openclaw, right? Mult bot/claude bot/openclaw. I use a agent tool from a company called Noose Research called Hermes. But the idea is it's similar to Siri with a big difference. It remembers your conversations, your prior conversations. It knows about you. It has some background about what you're looking for, what you want. It can even, it can even joke around with you based on. It does with me anyway.
Leo Laporte [01:42:54]:
Based on what it knows about me. You know, when I, when I have, I, you know, I record my meals in it. I say, hey, I call it Quicksilver. Hey, Quicksilver. I just had a sandwich and a, and a can of Coke. And it will say things like, that's a lot of carbs, Leo. You better, you better eat a salad for dinner tonight. That experience is pretty satisfying.
Leo Laporte [01:43:18]:
If Siri can become more of a personal assistant, I think that there's. And of course, yes, Sam, it'll also have to be able to open your garage door. Or it's just a dumb bot. But if it can be more functional and be more personal, isn't that a big step toward what people want from AI or is that just me?
Gary Rivlin [01:43:39]:
No, personal assistant is the holy grail for a Lot of companies, the AI agent. I mean, 2025 was the year the AI agent and that didn't work. So now it's 2026, the year of the AI agent. And this stuff is hard. It finally is starting to have a pretty good memory.
Leo Laporte [01:43:55]:
I've had to put a lot of work into it. I mean, it doesn't happen automatically, that's for sure.
Gary Rivlin [01:43:59]:
And that's my point. You know, it's just, I think 2027 and 2028 are going to be the year of the AI agent because, you know, I mean, humans are resistant to change. This stuff is complicated. The technology isn't quite there yet. There's a whole list of, of reasons and you know, but just to get a little bit, you know, nerdy for a sec, like, you know, the history of AI was, you know, rules based. We'll just kind of through sheer muscle, just like give every alternative and like expert systems. Expert systems, Millions of, and millions of lines of code that was rejected. No machine learning.
Gary Rivlin [01:44:29]:
Just let it learn in the fashion of a human. Just give it a bunch of, you know, data, whether it's, you know, books, movies, whatever kind of thing. But I really think the answer is it's not an either or. You know, I think the next step is going to be something of a hybrid because like, yes, it's brilliant. It could, you know, tell you, you have a conversation with you, but can it open my garage door? And so like, you know, one version, one approach is really good at a set of things. The other approach is good to a different set of things. And you know, I'm kind of thinking that, you know, even though rules based expert systems were rejected, I do think they're going to be in our future again for exactly what Sam is talking about.
Sam Abuelsamid [01:45:08]:
And maybe I'm just getting old and cranky, but, you know, I don't need the assistant to be funny with me. I don't need it to be joking with me.
Leo Laporte [01:45:15]:
That's pretty much a plus.
Sam Abuelsamid [01:45:17]:
I just need it to do exactly what I ask it to do.
Leo Laporte [01:45:20]:
Alexa plus it doesn't. Basically there's a sassy mode which we turned on and all anybody ever uses Alexa for is, you know, set a timer for asparagus. Yeah. And Alexa plus will say, oh, that's going to be some crunchy experience or, or dinner sounds great. When can I come over? And it's no smarter. It's just sassier.
Sam Abuelsamid [01:45:42]:
Yeah, I mean, I, I don't, I want smart. Not. I don't care if it's. I don't care if it's sassy, if it's dumb. No, that's not helpful.
Leo Laporte [01:45:51]:
It isn't.
Molly White [01:45:52]:
I'm also. Maybe I'm old school for this, but I hate talking to computers, like, speaking to computers. And I don't know if I'm ever going to get over that.
Leo Laporte [01:46:00]:
You don't have carpal tunnel syndrome, obviously.
Molly White [01:46:03]:
No, yeah. No.
Leo Laporte [01:46:05]:
Yeah.
Gary Rivlin [01:46:06]:
I think you're in the minority on that, Molly. I think people, you know, once mid 2024, before you can speak to ChatGPT, I think that was kind of a big breakthrough for them. I mean, that's kind of the Star Trek vision. We just talk to the machine. Yeah, exactly. And it talks back.
Sam Abuelsamid [01:46:27]:
I don't think. I assume neither of you work in an office with other people. I don't either. I've worked remotely for a dozen years. But when I worked in an office with other people, and increasingly people are being called back to office full time. You don't want to be talking to your computer in an office. That is just not a good thing. That's a terrible experience.
Gary Rivlin [01:46:48]:
You say that, Sam and I 100% relate to what you're saying. But the big thing now was it whisper flow.
Leo Laporte [01:46:56]:
I got it going right here. Look at this. This is on my Mac. As we're speaking, it has been transcribing everything that you've been saying.
Gary Rivlin [01:47:06]:
And so what's going on is like the fellow who made Claude code spoke about this journey. Yeah, yeah, you got it. Instead of typing into a box where we're kind of thinking about it going back, trying to perfect it, he found that using whisper flow and just speaking the way humans speak, the machine gets so much more out of it. And so I saw this picture, some article where, where they showed a room full of computer encoders, all of them whispering to whisper. And, you know, I think it's terrible. Exactly. It's like the movie 1984, like, dystopian thing. But I actually had the vision, Sam.
Gary Rivlin [01:47:46]:
I think that in some Silicon Valley places now, given that this genius fellow has said it, that now everyone's going to be talking to their machines aloud.
Sam Abuelsamid [01:47:56]:
I am so glad that I'm old enough that I am probably never going to have another office job.
Molly White [01:48:00]:
Same. But I went to a conference actually, where there's this basically kid, I think, I mean, I think he was in college, but barely, who had created some sort of a device that was exactly for that purpose. And it was. The idea was that you didn't actually need to really vocalize you just sort
Leo Laporte [01:48:16]:
of mouthed, I think, oh, I saw that video. That looks so dopey.
Molly White [01:48:21]:
I don't know if it's the same guy or not, but it was an interesting idea at least, which is that, you know, maybe we wouldn't.
Leo Laporte [01:48:28]:
Is it lip reading or sub vocalization? It was a thing.
Molly White [01:48:31]:
No, it's. Yeah, it sort of attaches back here and. Yeah.
Sam Abuelsamid [01:48:35]:
Sort of a reverse bone conduction.
Molly White [01:48:36]:
Yeah, yeah,
Leo Laporte [01:48:40]:
there. I mean, people are doing that. There's a company that. That's has a chip that you wear in a hat that is reading your brain waves. They're not good enough yet, but I imagine at some point machine learning will get good enough so that it can turn those brainwaves in. Into letters and words.
Molly White [01:48:56]:
Then I won't even have to type. I can just think it.
Leo Laporte [01:48:58]:
You just think it.
Sam Abuelsamid [01:48:59]:
Well, there's a friend of mine, he's in a startup now called Numo, that they're developing a system that goes into the headrest of the car that measures the EEG waves to help detect when you're distracted or drowsy and detects a bunch of things about the driver to instead of watch. High end. Yeah. Well, and it works in conjunction with that. So it gives you more information than what you get from just looking at, you know, where you're looking. But it's actually brilliant. Yeah.
Leo Laporte [01:49:31]:
Because I. I get sleepy.
Molly White [01:49:32]:
These things always end up getting used as surveillance, which is what freaks me out.
Sam Abuelsamid [01:49:36]:
Yeah.
Leo Laporte [01:49:37]:
Well, your insurance is going to go up if your beta waves are too prominent.
Molly White [01:49:40]:
Right?
Leo Laporte [01:49:41]:
Exactly. Have to pay more insurance. Yeah, that's a little scary. Somebody.
Molly White [01:49:47]:
Oh.
Leo Laporte [01:49:47]:
Let's take a break and then I'm going to talk about the fake EULA in the blockchain. I'm sure you saw this story, Molly. It's hysterical. You're watching this Week in Tech with Gary Rivlin, Molly White, who knows a little bit about crypto, and Samuel Bass, Amit, who knows a little bit about cars. We have crypto and cars coming up. So this happened yesterday. May 30, 15:29 UTC. Somebody wrote a fake EULA into the blockchain.
Leo Laporte [01:50:17]:
The EULA read by downloading this op return. I'll have to ask Molly what an op return is. You hereby consent to unrestricted access by federal law enforcement agencies to your residents, digital devices and personal properties. Assets may be searched, seized, or redistributed without further notice. Signed Donald J. Trump, President of the Internet. So what? First of all, what's an opera turn?
Molly White [01:50:43]:
So the op return is basically this location where you can store arbitrary data in a bitcoin transaction.
Leo Laporte [01:50:50]:
Okay. And you can't help but download it. Right. Because when you're using blockchain, you download a copy of the blockchain, right?
Molly White [01:50:58]:
Yeah. So all the transaction data is in there. I mean, there's definitely, like, ways that people will prune out a lot of excess data. But, yeah, for the most part, everyone downloads.
Gary Rivlin [01:51:07]:
Yeah.
Leo Laporte [01:51:08]:
When I. When I use my wallet, the first time I use it on a machine, it downloads. I mean, what is it now? It's 100 giga. It's a huge amount. 100 gigabytes or something of the. Of the entire blockchain literally downloads every.
Molly White [01:51:20]:
You have to validate it from the
Leo Laporte [01:51:21]:
beginning, every single transaction since Satoshi Nakamoto pushed the start button to now. Now. Then you can prune it. After that, you can prune it, and it gets much smaller. But that's a lot to download. Including this EULA six blocks later, which means almost instantly, they posted, I revoke my previous statement. Signed Donald J. Trump, President of the Internet.
Leo Laporte [01:51:49]:
But here's the thing. Nothing ever dies in the blockchain. Block 951,728 and block 951,734 will live for all eternity. Be.
Molly White [01:52:00]:
Yep. Along with every other block that people inscribe.
Leo Laporte [01:52:03]:
There's a lot of other crap. There's a whole bunch of stuff.
Molly White [01:52:06]:
Uploaded the Constitution the other day. Really cost him like, 80 bucks to do it.
Leo Laporte [01:52:12]:
But see, that'll be a nice historical record. There is. I've told.
Gary Rivlin [01:52:17]:
There are other records of it.
Molly White [01:52:20]:
Well, actually, that was the funniest thing is I read what I think was basically like an AI generated crypto slot media article, and it was like, finally, people can read the Constitution on the blockchain. And I was like, how many people have been trying to do that?
Leo Laporte [01:52:31]:
But there is other, less salubrious stuff. There's. There's CSAM on the blockchain.
Molly White [01:52:36]:
Yeah, well, and there has been a long running. Maybe not long running, but for a couple of years, there's been an ongoing fight within the bitcoin core developer community about whether or not they should basically limit the opportunity. Because it used to be that there was a very small, small cap on how much data you could put into the op return field. And then they lifted the cap, and so now there's. You can put really long stuff in there.
Leo Laporte [01:53:02]:
It was limited to 83 bytes.
Molly White [01:53:04]:
Yeah, exactly. It was tiny. And. And so now people are making, like, Bitcoin NFTs basically because you can store so much data than you previously could. And one of the core developers is like really concerned about the possibility of bitcoin becoming a CSAM haven because there's no way to erase it. And. But yeah, there's this huge debate within the bitcoin community around whether or not they should cap it, whether they should fork Bitcoin so that, you know, there could be a version of it that doesn't have the cap. It's a mess.
Molly White [01:53:37]:
Wow.
Leo Laporte [01:53:39]:
So just one more wonderful little feature of bitcoin.
Molly White [01:53:45]:
Well, in many cryptocurrencies, I mean, a lot of, you know, especially it's blockchain.
Leo Laporte [01:53:49]:
That's how blockchain works.
Molly White [01:53:50]:
Any, any blockchain that has arbitrary data, which is most of them, has this same issue.
Leo Laporte [01:53:59]:
The whole premise of blockchain is it's a decentralized, unforgable ledger.
Molly White [01:54:04]:
Immutable.
Leo Laporte [01:54:05]:
Immutable ledger. And so that's kind of in the nature of. Does become a little bit unmanageable when you get to hundreds of gigabytes.
Molly White [01:54:16]:
Yeah, I mean, I'm not sure how much Satoshi Nakamoto was thinking on like century scale when he designed.
Leo Laporte [01:54:25]:
Does it bog down because it's so big? No, it doesn't slow it down.
Molly White [01:54:28]:
I mean, it was slow to begin with, but for different reasons.
Leo Laporte [01:54:31]:
Right. The proof of work is slow all by itself.
Gary Rivlin [01:54:37]:
So the question I have is, so Trump comes in, he's pro crypto, their policies are going to buy bitcoin, whatever, for the reserves. I would have figured that for a four year period, the price of bitcoin, any cryptocurrency would go up and it did for whatever few months and then it just plummeted. Like why? I mean, the administration is so favorable, its policies are so favorable. They've, they've backed it, they're legitimizing it. Presumably they're buying up a bunch for, you know, our reserve. Why would it have fallen down like the way it did? I'm curious what you think.
Molly White [01:55:14]:
Well, for one, they're not buying it for the reserve. So there isn't that initial demand. There's. They haven't been able to. They would need to pass a law to actually authorize spending to purchase bitcoin for the reserve.
Leo Laporte [01:55:26]:
What's in the reserve right now is just confiscated by law enforcement.
Gary Rivlin [01:55:29]:
Right?
Molly White [01:55:30]:
Yeah. So there's no basically purchasing of bitcoin going on. But yeah, I mean, I think you're right. Like there was the Trump pump where, you know, it broke a new all time high shortly after Trump's inauguration. And, you know, it was doing well for a while. But ultimately, you know, bitcoin and other crypto assets are high risk, volatile assets. And we've seen so much macroeconomic economic instability coming largely from the Trump administration that people don't want to hold Bitcoin when they think that, you know, there's going to be huge tariffs or when there's a war with Iran or, you know, whatever it might be. And you see people following those same behaviors, which is that if they're worried about, you know, the economy in general, they don't want to be holding bitcoin.
Molly White [01:56:16]:
They want to be holding dollars or gold or whatever it might be. And, you know, there's that effect on the bitcoin price. But I think also people have a concern that, and especially increasing now that Trump is essentially damaging crypto.
Leo Laporte [01:56:33]:
Oh, interesting.
Molly White [01:56:34]:
Because his, you know, now so many people think crypto, they think, oh, that's that grift that the President of the United States is all tied up with.
Leo Laporte [01:56:42]:
And I think ransomware is also harming the.
Molly White [01:56:46]:
Well, certainly. I mean, and it always has been. Right. That's not new.
Leo Laporte [01:56:50]:
Ransomware was basically empowered by crypto before.
Molly White [01:56:55]:
Yeah.
Leo Laporte [01:56:55]:
I mean, literally, cryptocurrency, you had to buy cards at 711 to pay off your ransomware.
Molly White [01:57:01]:
Yeah. I mean, it's a rounding error of ransomware that doesn't use cryptocurrencies.
Leo Laporte [01:57:05]:
Right.
Gary Rivlin [01:57:05]:
So this case for crypto, to me,
Leo Laporte [01:57:07]:
at its peak, Bitcoin was $125,000. It's off 40% off. Its peak is now $73,000. So it's. It is. Do you think so? And there were people like the Winklevoss twins who said, oh, it's going to reach a million.
Molly White [01:57:22]:
They still say that.
Gary Rivlin [01:57:23]:
Yeah.
Leo Laporte [01:57:24]:
Well, yeah, because they have quite a bit of it.
Molly White [01:57:26]:
That's the thing, is the people who say that tend to have sort of a vested interest in it doing that.
Leo Laporte [01:57:31]:
They're hoping it hits a million.
Molly White [01:57:32]:
Yeah.
Leo Laporte [01:57:34]:
On the other hand, if you. If you bought it when it was a buck or 10 bucks or 1,000 bucks, you're still way ahead.
Molly White [01:57:40]:
Like the Winklevoss wins.
Leo Laporte [01:57:41]:
Like the Winkleville. Yeah. The problem I always had with it is, unlike almost any other investment, it's not tied to performance of anything. The value is tied to perception, nothing else.
Molly White [01:57:56]:
I would say that's not unlike everything else. I mean, if you look at the
Sam Abuelsamid [01:57:59]:
stock market, any fiat currency.
Molly White [01:58:02]:
No way different for fiat currency because you have the weight of the United States military I guess you're right.
Leo Laporte [01:58:07]:
The dollar is worth only what people say it's worth. I guess you're right. But. But if you buy a share of Apple, it's to some degree tied to Apple's performance.
Molly White [01:58:16]:
But what if you buy only to
Sam Abuelsamid [01:58:17]:
the degree that investors think it is?
Molly White [01:58:19]:
Yeah. Like, what about GameStop or Bed, Bath and Beyond or Tesla? Yeah, I mean, I would say that, you know, increasingly we're getting more. More and more financial instruments that are just. They're literally purely speculative, whether it's a meme coin, a more traditional type of cryptocurrency, a stock.
Leo Laporte [01:58:38]:
I tricked y', all, because I've been arguing this for years, and everybody yells at me when I say it, that stocks are just as. Just as imaginary as anything else. So I tricked you into agreeing with me. Thank you very much. Not that. Not that I can. You know, briefly, I thought I better get out of the stock market, because this is it. I mean, I have to live on this for the rest of my life.
Leo Laporte [01:59:01]:
That's my retirement savings. And I thought, well, what. What if I just have cash? Because then I know how much I'll have? But then inflation started to whittle it away, and I thought, I'm. And then I looked at the stock market, and unaccountably, it's been going up, and I thought, well, I guess I'm making a mistake not being. Even if it's imaginary, you kind of have to play the game.
Gary Rivlin [01:59:22]:
Yeah, I'm with you. The same set of reasons that Molly gave for why bitcoin crypto has not just soared. Tariffs, wars, instability, all that kind of stuff, and yet we just set another record, I believe, for.
Leo Laporte [01:59:39]:
It's amazing, the stock market. Yeah. So I wanted to ask you about this also, Molly. Hundreds of prolific Wikipedia editors are threatening to go on to strike. What happened?
Molly White [01:59:51]:
Yeah, so.
Leo Laporte [01:59:54]:
Because I should give you the. The audience, the information that you are one of those prolific Wikipedia editors, since you were a teenager, you've been editing Wikipedia. Yes, although I'm trying to go on
Molly White [02:00:08]:
strike, but that's largely because I've been busy. I haven't been doing it anyway.
Leo Laporte [02:00:11]:
Okay. You're not one of the people threatening to strike?
Molly White [02:00:14]:
Not at the current moment, but yeah. So I think basically what happened is there's something called the Community Wish List, which is a way for community members to basically request software features from the Wikimedia Foundation. And as a little bit of backstory, the community members are the people who edit the encyclopedia, write the content, all that kind of thing.
Leo Laporte [02:00:37]:
Volunteers, for the most part, right?
Molly White [02:00:38]:
Yes. Correct. The Wikimedia foundation is a foundation that employees employs software engineers and many other types of employees. And they, among other Things, developed the MediaWiki software that the Wikipedia and other wikis like it run on. And so the community wish list was a way for them, for community members to request that the software engineers at the foundation prioritize various features that the community felt were important.
Leo Laporte [02:01:05]:
And those engineers are paid staff, correct?
Molly White [02:01:08]:
Their staff members, yes, there are volunteer community developers, but that's a whole nother can of worms and not particularly pertinent to this. And so what happened recently was that the Wikimedia foundation announced that they would be changing the way that they handle the community wish list. And instead of having a team that was largely devoted to that task, they would instead sort of spread the work amongst the foundation's existing software engineers working on various other projects. And in doing so, they would also lay off a number of the engineers who were on that team.
Leo Laporte [02:01:44]:
Suman, who was writing about this, deputy chief product and Technology Officer at Wikimedia foundation, said it's a bottleneck to have this centralized team.
Molly White [02:01:52]:
That was what. Yeah, they were sort of portraying it as, you know, this isn't working very well. We hear feedback from the community that there's not enough priority going to this wish list. And so we're changing how it works because we think it's going to work better for. For you. But the community really took it more as we're getting rid of the people who do the wish list stuff. And all these other engineers who have their own list of tasks that they're trying to do already are not going to prioritize the community wish list tasks. And so it's going to only make the problem worse.
Molly White [02:02:19]:
And then there is the added fact that some people at least are. There's been an effort among Wikimedia foundation employees to unionize. And they're some of the people who, who were laid off were active in the unionization efforts. And so there's been allegations at least, that the layoffs were more to do with union busting than to do with any particular community wish list material. And so that's where a lot of the anger is coming from, is not just the fact that they feel like the wish list is being deprioritized, but they feel that the Wikimedia foundation was also punishing people who are trying to unionize.
Leo Laporte [02:02:58]:
The foundation has responded. Selena Duckelman, who's the CPTO of the foundation, has responded and Thanked people for the conversations. Do you feel like, I don't know if you followed this closely, maybe this is something you haven't been following. But was her response sufficient, you think?
Molly White [02:03:15]:
No, it wasn't sufficient. Certainly not to the people in the community. And I mean, I think one, one bit of background that's a little challenging for people who are not heavily involved with the Wikimedia projects is that there is a long running issue between the Wikimedia community, the volunteers and the foundation that has been a very strained relationship for decades. It's, I mean, almost an adversarial relationship and it's one that I think the Wikimedia foundation is working to improve. But incidents like this happen fairly often where the community feels like they're not being heard, like they're being deprioritized. And you know, in this case, I think people from the Wikimedia foundation made comments along the lines of like, this is us listening to you, we're trying to do. And you know, the community is saying, we didn't ask for this, we didn't want you to lay off these people. You know, this is the opposite of what we want.
Molly White [02:04:10]:
And so I think it was a miss there. I mean, I'm actually on a group, or I'm a member of a group that is specifically supposed to be improving communication between sort of the technical staff at the Wikimedia foundation and the community. And we were barely consulted and sort of not brought into the loop. So I think it was just a really bad situation all around. And another in a long list of examples where the Wikimedia foundation has not handled well its relationship with the community of volunteers that it relies upon to continue existing.
Leo Laporte [02:04:45]:
This is historic tension. I mean, you've seen it again and again in so many areas. Do you think the Wikimedia foundation is not a good steward of Wikipedia?
Molly White [02:04:57]:
No, I actually think the Wikimedia foundation is a good steward, but I think that it is unfortunately prone to stepping on rakes when it comes to, to its relationship with the community. I mean, it, despite having been the steward of Wikimedia for so long, it seems to have a very poor understanding of the Wikimedia community and how to interact with that community and to give, I mean, to be fair, the community is very challenging and I've been on the opposite side of it before as well. But you know, this is, I think it's just frustrating to see things like this happening over and over and over again. And, and I should Add that some of the people who are laid off were. They started as community members, they were community developers, they had some of the best context available as far as developing some of the features on the wish list. And so to see them removed and then have that painted as. This is us prioritizing the wish list. We're trying to make things better for all of you, like you asked.
Molly White [02:05:56]:
I think it felt like a slap in the face.
Leo Laporte [02:05:59]:
They did say that all six are currently still employed and that they've been trying to place them elsewhere in the organization. And three people have been offered jobs. So maybe there will be a happy outcome.
Molly White [02:06:13]:
Well, we'll see. I mean, I think. I think there's a lot of hope that the community wish list will continue to be maintained. I have my doubts that the strategy of trying to just have every engineer try to remember to work on it once in a while is actually going to, you know, cause it to be prioritized. But we'll see.
Leo Laporte [02:06:31]:
It's good to have a dedicated team. I agree. Yeah. Well, good. Well, thank you. I mean, I donate to Wikimedia every month. I. I feel like the Wikipedia is such a huge, valuable thing.
Leo Laporte [02:06:43]:
I'm really grateful to volunteers like you, Molly, who work so hard to keep it useful. And even in this age of AI and disinformation and truthiness, it remains still an amazing, useful tool and full of facts. And in fact, frankly, most AI would not be as good without Wikipedia. I know Google wouldn't be as good without Wikipedia.
Molly White [02:07:10]:
I don't think any of them would. It's a core part of the training set.
Leo Laporte [02:07:13]:
Right.
Molly White [02:07:13]:
And I think that's a big portion of why the conversation around a strike among editors is somewhat controversial. Because ultimately editors, you know, the top priority is to provide high quality information. And so doing something that would stop, that is, you know, even if. If editors think it would potentially further their aims, you know, it. It doesn't feel good to do that. And it's why things like blacking out Wikipedia like we did, I think a decade ago for sopa. Yeah, that was enormously controversial. And this type of a thing is as well.
Leo Laporte [02:07:52]:
We blacked out twit as well. In fact, we did all our shows in black and white that day for. Yeah, yeah. Okay. Well, I hope things continue. Does Wikipedia, Does Wikimedia do, you know, get any support from these frontier AI companies? I mean.
Molly White [02:08:13]:
Yes.
Leo Laporte [02:08:14]:
Oh, good.
Molly White [02:08:14]:
Yeah. So there is.
Leo Laporte [02:08:15]:
They should be donating like crazy.
Molly White [02:08:17]:
Yeah. There is basically a model called Wikimedia Enterprise where AI companies Pay for. They don't pay. It's not that they're paying for the data, they sort of pay for the highway. So it's, you know, a sort of a high access. Yeah. It's a high speed sort of API and it's, it's formatted in such a way that is much easier for these companies to ingest and they in turn, you know, pay for the, for the privilege. And so I think it's a fairly good system.
Molly White [02:08:49]:
It was, it was controversial when it was announced, I remember, because it was misunderstood.
Leo Laporte [02:08:53]:
Right. Doing deals with AI companies own, you know.
Molly White [02:08:56]:
Yeah, well, and I think people thought that, you know, that it was. There were some questionably written headlines, I think, around the announcement that almost implied that. That Wikipedia was incorporating AI content into Wikipedia. Yeah. And so I think people were upset about that. And then, I mean, some people, it was like, oh, you're doing a deal with the devil, which, you know, you can't really.
Leo Laporte [02:09:17]:
We had Jimmy Wales on right about that time to clarify things on intelligent Machines. I remember. Yeah, good, because it would be the greatest tragedy ever if anything were to happen to Wikipedia. It is, yeah.
Gary Rivlin [02:09:29]:
I mean, the big picture here, I mean, I've been around long enough to see it go from, you know, a small site into what it is now. Like, I remember literally in the mid-2000s, I was at the New York Times, A memo came out, I was sent out to all of us, all of the reporters, like, if you use Wikipedia as your source, it's a fireable offense.
Leo Laporte [02:09:48]:
Wow.
Gary Rivlin [02:09:49]:
And, you know, I mean, well, because
Leo Laporte [02:09:50]:
back then it's the same as the AI regulation today, it's the same rule.
Gary Rivlin [02:09:55]:
And, you know, and back then I kind of got it that, you know, there was a lot of revenge. You didn't like someone, so you wrote up the entry. And like, I've just been so impressed. Like, when I want to get smart about something, it's so good.
Leo Laporte [02:10:07]:
What a resource.
Gary Rivlin [02:10:08]:
The history, why it works, how it works, that, that kind of, that kind of stuff. But it really kind of. I guess the shame, I mean, listening to this, I knew nothing about this, is that, you know, I feel like in the mid 2000s, I couldn't have told you Wikimedia Wikipedia was going to in quotes work or not. Will it would it be? Essentially no one knew.
Molly White [02:10:28]:
Neither could we.
Leo Laporte [02:10:29]:
Yeah.
Gary Rivlin [02:10:30]:
And you know, and just they did such a great job of bringing in the volunteers and all this. So, yes, this is, this is credit
Leo Laporte [02:10:36]:
to people like Molly.
Sam Abuelsamid [02:10:38]:
It is an amazing resource.
Leo Laporte [02:10:39]:
Buckled down and made sure the content stays good.
Molly White [02:10:42]:
Well, and I mean credit to journalists like you guys because that's where it comes, that's where the information comes from, among other places.
Leo Laporte [02:10:49]:
But those volunteers who sit there, sit there on the discussion pages and you know, really protect the content. I mean, God bless him, that is, that is the most important work in the world. Especially in a world of AI filled with AI. Slop.
Molly White [02:11:01]:
Yep.
Leo Laporte [02:11:02]:
Speaking of slop.
Gary Rivlin [02:11:04]:
Gary.
Leo Laporte [02:11:04]:
What Are you ready? Now here's the luche. This is Ferrari's first ev. Johnny I've and love whatever. What is his love from. From weirdest name of a company ever designed this thing. Johnny said, oh, I love switches. Which is hysterical because he never seemed to love switches much when he was at.
Sam Abuelsamid [02:11:25]:
Well, was it him or was it Steve that hated.
Leo Laporte [02:11:27]:
Steve hated buttons. Yeah. People have commented on the design. The interior looks pretty sweet. I like the interior. The exterior maybe not so much.
Sam Abuelsamid [02:11:40]:
There's another good picture in the discord of the, of the luche being charged. So I think, you know, the problem, the fundamental problem with the Johnny, Johnny
Leo Laporte [02:11:52]:
always wanted to charge everything from the bottom.
Sam Abuelsamid [02:11:54]:
So that was the fundamental problem with the Luche is not necessarily so much the design itself, it's just that it's not, it doesn't fit as a Ferrari.
Leo Laporte [02:12:06]:
It doesn't look, it makes no sense. Ferrari.
Sam Abuelsamid [02:12:09]:
Yeah, yeah. I mean a lot of people commented, you know, this, this would have been a great design for something like, you know, Honda's now canceled Zero Series EVs. You know, it's, it's also got a lot of resemblance to a Jaguar. I pace. There's, there's, you know, a lot of things that it looks like, but it doesn't, it doesn't look like what, what they think a Ferrari should look like. Yeah, and you know, the executives from Ferrari have been talking this week about it, you know, and you know, the way they explain it is that, you know, we, we still have all the other traditional Ferraris, the cars that Ferrari lovers expect. But Ferrari, you know, Ferrari owners tend to own multiple cars. They don't just own a Ferrari.
Sam Abuelsamid [02:12:54]:
They, they own many different kinds of vehicles and some, sometimes they want a sedan like this. You know, I mean if you, if you look across the industry at premium brands, you know, Porsche really got this kicked off 25ish years, launched the Cayenne, you know, their first SUV and the, the Cayenne. You know, a lot of Porsche fans still dislike the Cayenne, but the fact is the Cayenne has been the best selling Porsche ever since it Launched.
Leo Laporte [02:13:26]:
Wow.
Sam Abuelsamid [02:13:26]:
And if, if it wasn't for the poor, if it wasn't for the Porsche Cayenne and then subsequently the, the Maan and, and the, the Taycon, which is, well, not so much the Tayon, but the, the, the Panamera. If it wasn't for those vehicles, then Porsche probably would not be in business. Still building 911s today. They could not survive just on building sports cars. You know, same thing is true with Lamborghini. They, you know, they have the Urus, Rolls Royce, Ferrari Cullinan. A few years.
Leo Laporte [02:13:57]:
Like just a car for everybody.
Sam Abuelsamid [02:13:59]:
No, no, but, but you know, they, they, I think they feel like they need, or they wanted some other option, you know, and they, they were looking for some way to get into the EV business. Now Grant, I am not defending this particular car. I still think, you know, I don't, I don't think it was necessarily a bad idea to do a Ferrari sedan. But this, I don't think that this is a good design for a Ferrari. It could have been for any number of other brands, but not as a Ferrari. So that's, that's where the, the problem is. But I do like the interior. I think the interior has got some really interesting analog clock for one thing.
Leo Laporte [02:14:41]:
Yeah.
Sam Abuelsamid [02:14:41]:
And you know, the, the touchscreen that, you know, is on a pivot, you know, so you can pivot it over towards the, the passenger. You know, if they want to do something on the touchscreen, you got manual vents, you know that you can just reach out and grab and redirect the airflow.
Leo Laporte [02:14:56]:
Unlike, unlike the BMW, which has.
Sam Abuelsamid [02:14:58]:
Yes.
Leo Laporte [02:14:59]:
A touchscreen for fence.
Sam Abuelsamid [02:15:00]:
Yeah, yeah.
Leo Laporte [02:15:02]:
Tesla.
Sam Abuelsamid [02:15:02]:
Yeah. Yeah. There's, you know, there's obviously been a lot of speculation in the last, the last week that this is actually, this originated as the design for the Apple car.
Leo Laporte [02:15:16]:
Well, that's what the question was for me and I'd seen that is that this is what Johnny I've wanted the Project Titan Apple car to look like. You think that's true?
Sam Abuelsamid [02:15:25]:
I wouldn't be surprised if it would have been similar to this shape. But you know, I think the interior
Leo Laporte [02:15:31]:
was going to be self driving. Right?
Sam Abuelsamid [02:15:33]:
Yeah, well, that's what that was. The goal was to make it, you know, self driving, to have no, have no human controls in it.
Leo Laporte [02:15:38]:
Right.
Sam Abuelsamid [02:15:39]:
You know, they, Apple was nowhere close to being able to achieve that.
Leo Laporte [02:15:43]:
Right. But that's why they killed it.
Sam Abuelsamid [02:15:45]:
Yeah, well, I mean there's a lot of, a lot of reasons why they killed it. You know, I think, think, you know, they, they, they probably realized that they were never going to get the kind of margins on a car that they get with the rest of their business. And you know, they figured if we can't get 35, 40% profit margins on the car, you know, and they're, you know, they're certainly not going to be able to sell very many, you know, six hundred thousand dollar cars. So if they can't get the profit margins they want, then, you know, they, they killed it. So, you know, and I think that was probably a wise choice, you know, for Apple. Apple doesn't need to be in the car business. But you know, Ferrari I think could have, could have done something more Ferrari like and still achieved the overall goals of this particular car.
Leo Laporte [02:16:35]:
Well, Jony, I've has many jobs, one of which is for which he got paid $3.2 billion, is designing a device for OpenAI. We don't know what will be, what it will look like, but there's some rumor that it might be out by next year. A pendant perhaps? I don't know. We'll see.
Gary Rivlin [02:16:56]:
Yeah, that would be my guess.
Leo Laporte [02:16:57]:
You know, kind of not glasses. Seems like the rest of the world's kind of converging on the idea that glasses are going to be the AI interface.
Molly White [02:17:06]:
But people get creeped out by glasses, especially if it has a camera they think you're recording.
Sam Abuelsamid [02:17:11]:
Well, would they be any less creeped out by a pendant that had a camera on it?
Molly White [02:17:14]:
That's true. If it has a camera, it's kind of the same deal.
Sam Abuelsamid [02:17:17]:
I did actually see somebody with one of those Humane AI pendants on a plane once. Yeah, on an airplane a couple of weeks before Humane AI went poof.
Leo Laporte [02:17:28]:
Yeah, apparently Apple's glasses, which they're calling or going to call the audio, Apple Audio. Even though they will have cameras, I think they don't want anybody to think about the cameras. You kind of need cameras because part of this is the AI seeing what you're seeing and informing you about it. It's going to be interesting. Apple's going to try this. I think that's the rumor.
Gary Rivlin [02:17:53]:
It's funny, I'm more creeped out by the audio. It's like wearing a pen and it's always listening.
Leo Laporte [02:18:00]:
Oh, recording.
Gary Rivlin [02:18:01]:
Yeah. I mean, do I have to get.
Sam Abuelsamid [02:18:04]:
You don't want to hang around with Leo then.
Leo Laporte [02:18:06]:
Well, no, I stopped wearing all those people yelled at me. I used to wear the B bracelet and I had all the recording devices because that's part of the personal agent thing. The ideal personal agent would hear everything I Am hearing. See everything I'm seeing. Digest it, absorb it and offer it for my consumption later or for my queries down the road. What were Molly White and I talking about with Wikimedia? What was that all about, Leo? Don't you remember? No.
Sam Abuelsamid [02:18:37]:
That's why I asked the question.
Leo Laporte [02:18:39]:
I'm asking you. Don't you remember? I have no idea, Leo. I don't know.
Gary Rivlin [02:18:45]:
I'm not sure if this. I'm not sure if this is great for marriages or the worst thing like that.
Leo Laporte [02:18:49]:
The reason I wasn't.
Gary Rivlin [02:18:50]:
You said this. No, I said that.
Leo Laporte [02:18:51]:
Yeah. No, my wife told me not to wear it. That's exactly.
Gary Rivlin [02:18:54]:
Okay, okay. It's been a lot of arguments, but I'm not sure that's a good thing.
Leo Laporte [02:18:59]:
Gerald TopPkins in our YouTube chat says that's because your wife's job is to remind you your personal agent is your wife. Oh, yeah, that's right.
Sam Abuelsamid [02:19:08]:
Oh, boy.
Gary Rivlin [02:19:09]:
Okay, don't touch that one.
Leo Laporte [02:19:12]:
She is not available at all during all conversations. All right, one more. One more break and then a couple of just wrap it up stories, including Peter Thiel evacuating the United States. Oh, boy. Good riddance. Oh, boy. More to come. You're watching this week in tech with Molly White, Sam Abul Samid and Gary Rivlin's great panel.
Leo Laporte [02:19:34]:
Great to have all three of you here. Couple of real quickies before we wrap things up. Peter Thiel, who has invested a lot into companies like Palantir, has decided it's not safe to be in the United States anymore. He has moved to Argentina, according to the New York Times, partly motivated by concerns about the future of the United States and shared beliefs with Argentina's right wing leader. There he goes into the presidential palace in Argentina. This is so annoying. Thanks for screwing everything up, Peter, and then leaving.
Gary Rivlin [02:20:12]:
So didn't he bought this huge compound in New Zealand. He became a New Zealand citizen. Is like many hundreds of acres.
Leo Laporte [02:20:20]:
Yep.
Gary Rivlin [02:20:20]:
And you know, I mean, I don't know from their government.
Leo Laporte [02:20:23]:
He has citizenship in New Zealand. He got that in 2011. He applied for a Maltese passport in 2022. Basically, this is what these rich guys do now. They have plan B, C and D. Yeah.
Molly White [02:20:38]:
Whenever I see someone accumulating passports, I'm like, are they trying to flee the country at some point point to like a non extradition country? Like you see it with the crypto billionaires. A lot of the time I'm always like, where are you trying to end up?
Leo Laporte [02:20:51]:
The New York Times says he's moved himself and his family to Argentina. His children are enrolled in a local school. The Argentine government is offering him permanent residence or even a citizenship. He's bought properties in not only in the Beverly Hills of. Of Buenos Aires, but also in neighboring Uruguay.
Molly White [02:21:16]:
He's just covering all his bases.
Sam Abuelsamid [02:21:17]:
You know who else ran off to Argentina and Uruguay in 1945?
Gary Rivlin [02:21:22]:
Right?
Leo Laporte [02:21:24]:
Interesting.
Molly White [02:21:25]:
I will say both of those countries do have extradition treaties with the United States, so maybe not. What's happening?
Leo Laporte [02:21:31]:
Yeah, I don't think he's fleeing prosecution. I think he's fleeing the. The economy that he built. Yeah, well, the surveillance.
Sam Abuelsamid [02:21:40]:
But he doesn't need to flee when you're that rich.
Leo Laporte [02:21:43]:
Of all people. I know.
Molly White [02:21:44]:
Well, have you read Douglas Rushkoff's book about that? About how like all these billionaires who are ruining the world are themselves building bunkers because they sort of fear what they've created.
Leo Laporte [02:21:55]:
We had Douglas on to talk about that book. It's a great book and it talks about, you know, the, the former anti ballistic missile sites and the ICBM sites billionaires are building little hidey holes in.
Gary Rivlin [02:22:09]:
Yeah.
Molly White [02:22:09]:
Under the Ballroom.
Leo Laporte [02:22:11]:
Oh, yeah. The Ballroom would make an excellent place to put many, many stories. I mean, it's just. It's just wild Survival of the Richest is the name of that book, by the way.
Gary Rivlin [02:22:27]:
You know, I kind of knew Thiel for a while. I wrote a profile. I profiled him in the New York times magazine in 2005, before he was such a big deal. I remember we were talking in the 2010s and he was talking about New Zealand and leaving, like, you know, you can't. Like, we were talking about Trump and just like, yeah, I don't know how it's gonna go, but it's such a broken system. Let's twist the kaleidoscope. It's like, okay, dude, like, I have kids, I earn. I can't just do that now.
Gary Rivlin [02:22:56]:
I guess he has kids, too, and he's going to run.
Leo Laporte [02:22:58]:
Doesn't matter. He can afford to do it. Right.
Gary Rivlin [02:23:00]:
Well, but the stakes in the experimentation, I think, are different. When it's just for you, you just kind of. It almost reminds me of Trading Places for a dollar. Bet you could screw up all these people's lives kind of thing because you're immune, you're protected from the consequences. But I guess the news from that is like, peter Thiel has kids. I did not know that.
Sam Abuelsamid [02:23:22]:
Yeah, I mean, I would have, you know, no problem with, you know, just loading all these guys on a starship and sending them to Mars, you know, they all seem to want to go there anyway.
Gary Rivlin [02:23:31]:
So Musk will get richer. He'll reach his million faster. Yeah, there you go.
Leo Laporte [02:23:36]:
So frustrating that they create this dystopia.
Molly White [02:23:40]:
I will say I did love the part about the Pope quote, not quoting, but sort of alluding to Tolkien in his AI thing where when Peter Thiel names all his companies after Lord of the Rings, volunteer and volunteer, it just felt like a little bit of a shady choice.
Leo Laporte [02:23:59]:
I think the Pope knew exactly what he has to do, is what you're saying.
Sam Abuelsamid [02:24:04]:
That was not a coincidence?
Molly White [02:24:05]:
No, absolutely not.
Leo Laporte [02:24:06]:
I think you're. Robo taxis are spreading across the US says the Wall Street Journal. And so is the backlash. You've seen this, Sam, I'm sure.
Sam Abuelsamid [02:24:21]:
Yeah, they're, they're having a few issues. Waymo, as they, as they keep scaling to more and more cities are encountering more and more problems. Recent weeks, you know, following some storms in the south, you know, you had multiple Waymo Robo taxis driving into flood waters and getting stuck with passengers.
Leo Laporte [02:24:44]:
This is Atlanta cul de sac.
Sam Abuelsamid [02:24:47]:
Yeah.
Leo Laporte [02:24:47]:
For some reason all these Whamos have
Sam Abuelsamid [02:24:49]:
decided to basically repeating what they did in San Francisco. Yeah, well, there was another, there was another neighborhood where there was just strings of them just going round and round this roundabout in a neighborhood.
Molly White [02:25:01]:
It's like the ant thing where they
Sam Abuelsamid [02:25:02]:
get stuck, stuck in the spiral, basically. Yeah. They're still having some issues with recognizing school buses and responding properly to school.
Leo Laporte [02:25:13]:
That's not good.
Sam Abuelsamid [02:25:15]:
There was, there was at least one incident in California where one went speeding through a construction zone. So they have, they have paused all highway operations, you know, which they've been doing for, for a while now. So they, they stopped driving on the highway and they have reduced the, some of the operations in a variety of cities under, you know, under various conditions while they try to sort out these problems. But part of the problem here is, you know, over the last eight, nine years or so, Waymo, like everybody else, has transitioned to a more, a more and more AI based approach for their software stack, which actually makes it much more difficult to diagnose what's going wrong. You know, so they've gotten away from the rules based system that they started with and they're, it's much more of a monolithic AI.
Leo Laporte [02:26:11]:
Should we be worried? Morgan Stanley says 30 of the rideshare industry will be autonomous by 2032. Or is that, is that maybe a little.
Sam Abuelsamid [02:26:23]:
Yeah, I mean, well, you know, I mean it's, it's a big industry, you know, and it's, it's not going to entirely replace the rideshare industry anytime in the foreseeable future. Because the problem is the costs are still high and unless you get a really high level of utilization, the, the economics don't really work with these.
Leo Laporte [02:26:43]:
So they're currently still losing money then?
Sam Abuelsamid [02:26:45]:
Oh yeah, they're every, nobody is making money on this and they probably won't be for a number of years. Which is why companies like Ford and GM pulled out and said, yeah, we're, we're not going to get into the robo taxi business. We got, we got too many other fish to fry right now. We don't need to be spending our money on this. So, you know what, what you'll see happening in more and more cities is that the robo taxis will kind of form the baseline, you know, because you, you have a real fluctuation in demand through the course of the day. It's not constant through the day, through the course of a day. So robo taxis will kind of form the baseline. And then during heavy demand periods, that's when you'll see the human driven cars coming in to supplement and crank up the supply of vehicles.
Leo Laporte [02:27:30]:
Here's something for you, Gary. There is a company called SHIFT that is offering free house cleaning in New York City.
Gary Rivlin [02:27:40]:
Train their robots. You have free, but you have to allow, you have to be observed. I am not signing up.
Molly White [02:27:46]:
Risk.
Gary Rivlin [02:27:47]:
Exactly.
Leo Laporte [02:27:48]:
We record first person cleaning footage to help train the next generation of household robots. So for a limited time, we cover the cost of professional cleaners. They come in, they have cameras attached to their head, they record first person video, and then you get a spotless apartment. We get training data, everybody wins, says shift. You know, I could do it, huh?
Gary Rivlin [02:28:13]:
Yeah, I'm not going to do it. I, I, you know, almost the same reaction I do to cars is like, you know, 35,000, 40,000Americans die every year in cars. I think that autonomous vehicles, you know, could be safer. I'm all for, you know, what was from the Jetsons. Janie. Rosie, Rosie. You know, the idea that, you know, some robots could clean up like, it just, it's going to cost tens of thousands of dollars to buy the robot that's going to clean up or to enlist the service. So I don't know, this seems like a very long play.
Gary Rivlin [02:28:47]:
Let me put it that way. It'll be a long, long time before, you know, these, these bots will be cleaning up.
Molly White [02:28:55]:
Is this like a bipedal robot cleaning?
Leo Laporte [02:28:59]:
Well, we don't know because it's a human that comes in with the camera. Right.
Molly White [02:29:02]:
But do you know what they're training for?
Gary Rivlin [02:29:04]:
For?
Sam Abuelsamid [02:29:05]:
Well, yeah, I think to create bipedal robots because. Yeah, or to sell the training data to companies that are doing bipedal.
Leo Laporte [02:29:12]:
More likely it's that Sam that they, they know that, you know, Optimus Elon Musk's Optimus robots.
Sam Abuelsamid [02:29:17]:
Well, there's, there's the one, I can't remember the name of the company. Joanna Stern did a great video on this one a few months ago where she had one of these robots, one of these bipedal robots in her house. And it turned out that, you know, most of what it was doing was being, it was being tele operated. So there was somebody sitting back in Silicon Valley watching the screen crappy the view of, through the cameras of the, of the robot and tele operating this thing. And you know, so far none of these things are. The only ones that are even remotely impressive are the, the Boston Dynamics Atlas and a few of the Chinese robots.
Leo Laporte [02:29:55]:
The dog robots.
Sam Abuelsamid [02:29:56]:
Yeah, well, Atlas. Atlas is the bipedal one. Oh, it's Spot is the dog.
Leo Laporte [02:30:01]:
Oh, yeah. Okay.
Molly White [02:30:02]:
But I have recently come around to robot vacuums and I am a convert in that department.
Sam Abuelsamid [02:30:09]:
Even, Even with having a dog in your house?
Molly White [02:30:11]:
Oh, yeah, yeah, yeah.
Sam Abuelsamid [02:30:12]:
Okay.
Leo Laporte [02:30:13]:
Which one do you have?
Molly White [02:30:14]:
He's a little like afraid of it, but it's no big deal.
Leo Laporte [02:30:18]:
Which one do you use? Molly, do you want to give them.
Molly White [02:30:20]:
It's one of the dreams ones. I can't remember which model. Dreamy maybe. Yeah, my, my biggest thing like the vacuum, I could sort of take or leave, but it mops and I hate mopping so much that that has made a huge difference for me.
Leo Laporte [02:30:35]:
Oh, I, I entered dreamy and I got a picture of Cristiano Ronaldo.
Molly White [02:30:39]:
It's dream with an E. Oh, no, that's still.
Leo Laporte [02:30:41]:
It is him. It is them.
Molly White [02:30:46]:
Your thing for.
Sam Abuelsamid [02:30:46]:
These are the guys that had an event in San Francisco of about a month ago and they unveiled this rocket powered EV that would go 0 to 60 in like 0.9 seconds.
Gary Rivlin [02:30:57]:
Hell yeah.
Leo Laporte [02:30:58]:
Can I do that to my vacuum crash even faster. Ladies and gentlemen, I think we can wrap things up with that dystopian note. Ladies and gentlemen, thank you so much for being here. Gary Rivlin, whose house will not be cleaned by volunteers, is the author of a very good book about this which turned out to be kind of prophetic. AI Valley. You could find more on his website. Gary rivlin.com. when's your podcast? Are you starting a podcast, Gary?
Gary Rivlin [02:31:28]:
No, I'm not starting a podcast. I don't think there's enough of them, though. But I'm not doing my podcast.
Leo Laporte [02:31:33]:
No, we need more podcasts. Well, I just look forward to your novel or your son's screenplay. One or the other. One or the other. I think that's such a. I'm looking
Gary Rivlin [02:31:40]:
forward to him making 20 million in the first three days.
Leo Laporte [02:31:43]:
There you go. That's right. That's right. Tell them to find support.
Gary Rivlin [02:31:46]:
I want points. You did it in my apartment. I want points.
Leo Laporte [02:31:50]:
Gary, always a pleasure. Great to see you. Thank you for joining us. I appreciate it.
Gary Rivlin [02:31:53]:
My pleasure.
Leo Laporte [02:31:54]:
Molly White. You gotta go there. Citationneeded news. Great newsletter and podcast. If you look at the podcast feed, there will be a story about Citizens United coming soon.
Molly White [02:32:06]:
It'll be on the newsletter, too. You can't miss it.
Sam Abuelsamid [02:32:09]:
Nice.
Leo Laporte [02:32:10]:
Nice. And of course, all of our other sites as well. It's always a pleasure, Molly. Thank you for the work you do. It's, I think, so important. We appreciate it.
Molly White [02:32:18]:
Thanks for having me back.
Leo Laporte [02:32:19]:
Yeah, always a pleasure. And my car guy, Sam Abulsamed, who secretly feeds me car ideas. I have an AI Go out every week, Sam, and look at everything and try to. Because I've got to decide what to get in the next few months and. But I think your recommendations are 10 times better than the AI, so maybe. Would you mind if I can incorporate your brain into.
Sam Abuelsamid [02:32:42]:
Absolutely.
Leo Laporte [02:32:42]:
Okay. I think I will.
Sam Abuelsamid [02:32:45]:
And I would like to put in a plug. Starting next Sunday, my friends and I will be doing another Operation Frodo event.
Leo Laporte [02:32:56]:
What's that?
Sam Abuelsamid [02:32:57]:
This is.
Leo Laporte [02:32:57]:
Are you throwing a ring into a volcano?
Sam Abuelsamid [02:32:59]:
No, no, no. No rings, no volcanoes. This is dog rescue. So we've been doing this every December now for the last four years, and we transport rescue dogs from Nebraska to the Pacific Northwest. So it's. It's all people in automotive media. And so we borrow vehicles from automaker press fleets and we get donations to cover the costs. And so we will be leaving one week from today.
Sam Abuelsamid [02:33:32]:
Next Sunday morning, June 7th, we'll be leaving Omaha, driving to Cheyenne and. And then on to Salt Lake City, Ontario, Oregon, and finishing up in Portland with about 12 or 13 dogs this time. We've. So far we've transported 88 dogs over the last four years. And this time is going to be a little different because we're doing it all with a fleet of electric vehicles.
Leo Laporte [02:33:57]:
Oh, cool.
Sam Abuelsamid [02:33:58]:
So we've got a Cadillac Escalade IQ, a Lucid Gravity, a Hyundai Ioniq 9 and a Kia EV9.
Leo Laporte [02:34:04]:
And are the car makers providing those vehicles?
Sam Abuelsamid [02:34:07]:
Yeah, they're loaning us the vehicles EVGO is sponsoring. They're providing us with free charging at pilot flying J travel centers, which there's a whole string of them all the way along. I 80 enter. I mean there's, there's actually a lot, a lot more DC fast charging stations popping up this year even, even with the, the federal government trying to pull back and kill this stuff. There was a thousand new DC fast charging stations that launched in the first quarter of this year. So we will, I will, I will be driving that Cadillac from here from my home out to Omaha starting on Friday afternoon. So be going about 2700 miles.
Leo Laporte [02:34:50]:
Wow.
Sam Abuelsamid [02:34:50]:
To transport these dogs.
Leo Laporte [02:34:51]:
Where can we learn more about this year? This is last year's Operation Frodo.
Sam Abuelsamid [02:34:56]:
Yeah. So this is, this is the main site, animal rescue rigs. Okay. And they're, you know, you'll find stuff on here and you'll find on social media from, from all of the participants. So just look around on Instagram and Blue sky and, and elsewhere. Some, some people will be posting on Facebook and our, the longer term goal, we're, you know, raising a fund to purchase and equip vehicles to put with shelters and rescue organizations around the country so that they can transport animals whenever they need to, rather than just waiting on us to do it a couple of times a year.
Leo Laporte [02:35:38]:
Yeah.
Sam Abuelsamid [02:35:39]:
And so, you know, you can, there's a donate button on the site on animal rescue rigs. You can make donations there as we, as we build up this fund and continue rescuing dogs.
Leo Laporte [02:35:51]:
That is, that is so.
Sam Abuelsamid [02:35:52]:
And you can also, you can, yeah, you can also donate to some of the specific rescue organizations that we work
Leo Laporte [02:35:58]:
with car makers for doing this. That is.
Sam Abuelsamid [02:36:01]:
Yeah. We've had a whole bunch of automakers that have, that have supported us over the last four years. Subaru's always been a big supporter of.
Leo Laporte [02:36:09]:
Yes. Because everybody who has a Subaru has a dog. I've noticed that.
Sam Abuelsamid [02:36:12]:
But yeah, Toyota is, Toyota is a big supporter. Hyundai and Kia are really big supporters. They've helped us out on pretty much every one of these drives.
Leo Laporte [02:36:21]:
That's really cool.
Sam Abuelsamid [02:36:22]:
And this is the first time with GM and with lucid participating.
Leo Laporte [02:36:26]:
Oh, so what are you going to be driving?
Sam Abuelsamid [02:36:29]:
We, I'll be driving the Escalade. But then once we get to Omaha, we'll be over the four days of the trip, we will be rotating through all four vehicles that we've got. So I'll spend some More time in the Escalade. But also, is it all.
Leo Laporte [02:36:43]:
Is it all beagles? I see a lot of beagles.
Sam Abuelsamid [02:36:45]:
Yeah. So the. The organization in Omaha is the Bassett and Beagle Rescue of. Of the Heartland. So it's mostly beagles or beagle mixes. And the reason why. Why. Why there's a lot of beagles.
Sam Abuelsamid [02:36:58]:
Nick Miles, who started this whole thing four and a half years ago, one of his dogs, a beagle, died, and he got connected with Basset and Beagle Rescue of the Heartland, and he wanted to adopt one of their dogs. And he found out that there was a real surplus of beagles, especially in the fall in the Midwest, because what happened. What happens is hunters at the end of hunting season, the dogs that aren't doing the trick for them, they will just abandon them or shoot them. And then puppy mills that were, you know, if they don't sell the pumps by the time they're about four months old, they want to get rid of them. So they have. There's a surplus of them in that part of the country, but they need
Leo Laporte [02:37:42]:
them in Washington State.
Sam Abuelsamid [02:37:43]:
There's a demand for them in the Pacific Northwest. People. There's not. There's not very many of them that you can find, so we work with.
Leo Laporte [02:37:50]:
Drop one off in Petaluma on your way, would you? Because I love beagles. I would love a beagle. Yeah, we can get the.
Sam Abuelsamid [02:37:55]:
Probably make that happen.
Leo Laporte [02:37:56]:
Yeah, they got really good ears.
Sam Abuelsamid [02:37:58]:
Yeah.
Molly White [02:37:59]:
There's a remarkable sort of like, dog distribution systems.
Sam Abuelsamid [02:38:04]:
Yeah.
Molly White [02:38:04]:
They do the same thing from the south to the north. And.
Gary Rivlin [02:38:07]:
Yeah.
Sam Abuelsamid [02:38:07]:
I mean, with. With so much, you know, with so much of the bad stuff that we like what we've talked about today, something happened. It feels so good to do something positive, you know, and we all get. We get really attached to these dogs after four days.
Leo Laporte [02:38:19]:
I bet you do.
Sam Abuelsamid [02:38:20]:
But Robbie. Robbie Baldwin's going to be with us. He's going to be one of our drivers this time.
Leo Laporte [02:38:25]:
Wheel Bearings, your podcast. Yeah.
Sam Abuelsamid [02:38:27]:
So. So watch out for. Watch out for Operation Frodo on social media next week. Clyde. My friend Paul adopted Clyde in December. He fell in love with Clyde.
Leo Laporte [02:38:39]:
Clyde's adorable.
Sam Abuelsamid [02:38:41]:
Clyde is. He's a beagle and border collie mix.
Leo Laporte [02:38:46]:
Oh, boy. That's an interesting mix.
Sam Abuelsamid [02:38:49]:
Yeah, he's. He's a sweetie.
Leo Laporte [02:38:50]:
Interesting mix. Oh, look at the. These sweeties. I love that. Well, have fun in your drive. That is really, really cool.
Sam Abuelsamid [02:38:59]:
Yeah, I'm looking forward to it. Should be great. And. And, you know, we'll be writing stories about the. The trip and I'll be doing a story for greencars.com on road tripping across the country with EVs. So there'll be lots of interesting stories coming out of this over the next few weeks.
Leo Laporte [02:39:17]:
It's a great beagle migration.
Sam Abuelsamid [02:39:19]:
Yeah.
Leo Laporte [02:39:20]:
Thank you Sam. bullsamit. Of course. Wheelbearings Media is this podcast. Thanks to all three of you. Really great show. We really appreciate it. We do TWIT every Sunday afternoon and you know, I love doing this because I get to talk to smart people about what's going on in the world.
Leo Laporte [02:39:34]:
I hope you enjoy the show. You can watch us do it live if you want, from 2 to 5pm Pacific, 5 to 8 Eastern or 2100 UTC if you're in the club. Of course, club members get to watch special access behind the velvet rope in our club, Twit Discord. But we also stream it on YouTube, Twitch, X.com, facebook, LinkedIn and Kik. You can watch and chat with us from 2 to 5 Pacific every Sunday. After the fact, you can download the show. We have audio and video available at our website, Twit TV. There is a YouTube channel dedicated to this week in tech.
Leo Laporte [02:40:08]:
Great way to share clips. And of course the best way to get any of the shows on the Twit network is to subscribe. Just find your favorite podcast client, we're in all of them and search for Twit. Pick the shows you want. You get them automatically as soon as we're done. And if you like the show, leave us a good review because that helps spread the word. We've been doing this for a long time and I thank everybody for your support over these 21 years. A special thanks to our club members who really make this possible.
Leo Laporte [02:40:36]:
They cover the costs, advertising helps, but it's the club members that that get us to the to pay everybody who's working on the staff. People like Benito and Kevin King will be editing the show. We appreciate your support. TWiT TV Club TWiT. If you're not a member and you'd like to know more, thanks for joining us. We'll see you next time. And as I've been saying for 21 years, smile for the camera. Another Twit is in the can.
Leo Laporte [02:41:02]:
Take a picture of them right there. That one got it.
