Hard Fork
Hosted by Casey Newton and Kevin Roose
Introducing DJ TaskRabbit.
14,871 words · 13 speakers
Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
Support for this podcast comes from Dell Technologies. It's time to put AI to work. Is your business ready? The challenge isn't getting started with AI. It's scaling beyond pilots while maintaining control of your data, governance, infrastructure, and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner, Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.
I had an AI experience this week.
What was that?
I just started seeing a new doctor.
OK.
And I go to my first intake appointment with this doctor, and the doctor comes into the room, nice guy. He sets this little thing down on the table next to him. It's basically like a phone, but the screen is off, and it's got a little microphone attached to it. And I'm thinking, OK, he knows I'm a podcaster. He's trying to make me feel comfortable.
In case you have anything you want to say and want to kind of get it down, you can do it in the office.
Exactly. And then he tells me, this is my AI scribe.
A scribe?
Yes.
We haven't seen scribes as a job in this country for thousands of years.
Since medieval times.
Yeah.
No, but the scribes are the people who accompany doctors when they're meeting with patients and write down what people say.
So now, this job is being taken over by AI. And this doctor explained to me that he has this app that just sits on this little phone that's purpose-built for this kind of thing. And it will take notes on everything we say during the meeting, and it'll condense our notes down into a visit report and a summary, and it'll put it into my file. And like that, we'll save him tons of paperwork.
That's amazing. And sort of based on everything that it writes down about you, that's all fed into Facebook so that it can show you targeted ads on Instagram. Is that right?
I think HIPAA protects that, but we should check on that.
We should change that rule.
I'm Kevin Roos, a tech columnist at The New York Times.
I'm Casey Newton from Platformer.
And this is Hard Fork.
This week, Apple's got a brand-new mixed-reality headset, and Kevin and I tried it. Then, a dramatic hearing over child safety in the Senate. And finally, how a single-car accident took down the self-driving car company Cruise.
Casey, we went on a field trip last week.
We did, you know, as much as we enjoy being in this studio, this beautiful environment that we're in right now, it is awesome from time to time to leave the confines of the studio and go out into the world and get our hands on some new technology.
Yeah, so we were both invited down to Cupertino, California, to the headquarters of Apple to try out the Vision Pro, which goes on sale this week. We've talked about it on the show before. This is Apple's, what they're calling a spatial computing device, essentially a VR or an AR headset, although they don't use those terms. And it's attracting a lot of attention, first and foremost, because it costs $3,500 for the base model. But this is, I would say, the biggest hardware release of the year. And so I was very excited to go down to Cupertino and try it out myself.
Yeah, I was. Now, we should say, you know, a bunch of journalists did get their hands on this before. We did. You know, I would say that we were sort of brought in as the kind of cleanup crew among the journalists.
They were saving the best for last.
They saved the best for last. We were the last ones to get our hands on this. And, you know, I think we'd like to share some impressions. I would say that, you know, I think we have some positive things to say about it. And I know that sometimes when some of our listeners hear journalists saying positive things about technology, it sends them into a rage. So I just want to say sort of preemptively that we will also continue to criticize Apple. But, you know, sometimes people do cool things.
Yeah. So first, let's just talk about what this headset is.
Yeah, so Apple calls it a spatial computing device. And in many ways, they've set it up to be the eventual successor to the Mac computer. So that is sort of the level at which Apple is thinking about it. Of course, it's still very much in exploration mode. We don't know if it will get all the way there. But the Vision Pro is an effort to see, can you move away from laptops with keyboards and this sort of one physical display, put a helmet on your face, stay in it all day, have essentially infinite displays, manipulate objects with your fingers, navigate? The device with your eyes and invent a kind of new paradigm of computing. So this is like a really big swing, Kevin, right?
Like, Apple's had a lot of recent successes, whether it's the the watch, the phone, the iPad. You name it. Like, all of those are like, pretty big businesses. But I think all of them were easier to accomplish in their own ways than what Apple is trying to do with the Vision Pro.
Totally. It's a very ambitious kind of project. It's been many, many years and billions of dollars in R and D in the making. And it really felt like it, like it's a it's a beautiful device. It looks like a pair of ski goggles. It's not kind of bulky or clunky like other VR headsets that I've worn. It really looks like an Apple product and it has the price tag to match.
Yeah. And at the same time, Kevin, I think the most important question about this device, which I do think remains mostly unanswered, is what is it for or who is it for? Right? Somewhere in there, I think, is the real question about this device. And I think as Apple guided us through some demos, they tried to answer those questions for us.
And I think we should also just say as sort of a blanket caveat to this segment, like, this is not going to be a full review because we did not get a chance to really take this thing for a full test drive.
My demo was about 45 minutes. I think yours was about the same length. We were not allowed to actually take one home and test it in our own sort of home environment.
We couldn't throw it down a flight of stairs to see if it broke.
Exactly. And I would say it was a it was a heavily curated experience. Like, Apple definitely had a set of things that they wanted to show, at least me. And so it didn't feel like I actually got to sort of play around on my own terms.
Yeah, that's right. So these are impressions from a very guided, very curated look at this new device.
Yeah. So we went down, we, you know, I at least, it was interesting. They had us come in separately. It was like, like, we were too powerful if we if we had our demos at the same time. They didn't want us to combine forces.
That was actually by my request. I want to thank Apple for accommodating me on that one.
So at least my experience, I meet the Apple sort of minder at the gate. I get walked in, sort of past this manicured lawn, into this Steve Jobs theater, down a set of stairs, and into this like, what is essentially like a fake living room. So I sit down, I have basically two Apple employees in the room with me, one who's sort of giving me the demo, and one who's just kind of there keeping tabs on the situation. And the first thing you have to do is kind of do a little setup tutorial thing, to sort of calibrate it to your eyes.
They scan your face.
They scan your face, because there are no controllers with this thing. The way that you control your cursor essentially is just by looking around. So, once I got set up with the eye tracking and all the gestures, then Apple showed me these, what I thought was the highlight of the demo frankly, were these things, spatial photos and videos. Did you see these?
Yeah, this was really, really cool. So, you can either take these photos with your iPhone, or you can take them with the headset itself, although it seems like, in just about every case, you're better off taking these things with a newer iPhone.
Well, they only work on the iPhone 15 Pro and Pro Max.
Which is last year's iPhone. If you don't have last year's iPhone, this doesn't work.
Right. So, these are essentially 3D photos and videos, and you can view them in 3D in the Vision Pro itself. And, I don't know, have you spent a lot of time playing around with 3D photos and videos?
Uh, no, I mean, not really. I can remember going down to Facebook years ago, and having them show me 3D photos, and sort of telling me this was the future, and you would just kind of like tilt your phone around, and be able to shift the perspective a little bit. Not that impressive. On the Vision Pro, it does feel a little bit more of like a Black Mirror situation, where you're sitting inside somebody else's memory, particularly with these videos, like they showed me this spatial video of a family having breakfast, and of course, it was like a young family, and everybody was very adorable, and they're pouring orange juice, or whatever. And, you just like feel like you are there with them, and as I was looking at this, I thought like, I could see how this would be some family's cherished memory, and 20 years from now, the kids are all grown up, and they strap on a helmet, and they revisit this, and that that is like a special thing for them.
Yeah, I mean, this blew my frickin' mind, I'll be totally honest. I was like- Very ready to be skeptical because I have played around with a lot of 3D cameras in the past. I used to take 3D photos and videos just sort of like. Cause I wanted to like relive, I'm like a camera dad, you know, and like I have a kid and I just take a lot of videos. And so I've wanted something that feels a little bit more immersive.
I see.
And this thing is incredible. I mean, I saw the same demo you did. It sounds like there were a couple of photos and videos. There was, you know, what looked like a birthday party. There was a mom sort of making bubbles with her kid. And we should explain, it's not, these are not sort of like just projected at you in like a dark environment. You can see these kind of overlaid on the real world because of this pass through display that Apple has built. So I was in this living room, this fake living room, you know, with these Apple employees around me with sort of this coffee table in front of me. And this memory, this video just kind of popped up, you know, amidst all of that. And you're right, like it really did trick my brain into thinking that I was there in this scene.
It was, I've never seen anything like it. And like, I'll be honest, I got sort of a lump in my throat because I was picturing like capturing my son's like first steps this way and like revisiting it 20 years from now. And like, you know, that kind of thing I think will make this more compelling for especially parents.
Yeah, it is powerful. And I think particularly if you've already got a phone in your pocket and you can take videos like this, it winds up being pretty easy to take the video. And maybe if you don't even buy the first Vision Pro, which most people are absolutely not going to do, maybe three or four or five generations from now, you still have these, you know, videos saved in your iCloud somewhere and you're able to relive them again. So it does seem like an interesting development.
Right. So after the spatial photo and video demo, I got this demo of basically a movie theater experience where, you know, they showed me a clip from Super, did you get the Super Mario Brothers 3D?
Clip, and then a clip from Star Wars. And you could kind of transform your surroundings in the headset so that it looked like you were, you know, watching the Star Wars trailer on Tatooine, or you could be in a volcano in Hawaii watching. Super Mario Brothers in 3D.
Yeah, and an interesting aspect of this is that like the Apple Watch, the Vision Pro has this little control feature that Apple calls a digital crown, this little wheel that you can spin. And on the Vision Pro, you can dial the level of immersion up and down. So if you want to watch a movie while pretending that you're sitting in a volcano crater, you can sort of crank that dial all the way up and look all the way around you in 360 degrees and you see volcano crater. Or if you just kind of want like a hint of that, you can make it semi-transparent and the cameras in the Vision Pro will show you your surroundings in really high definition.
Yeah, and it actually works pretty well. They call this feature pass-through, which is a little bit of a misnomer for reasons that maybe we should take a minute to explain. It's not actually like, the display is not becoming transparent. It's just, it has cameras on the outside that are sort of capturing the room around you and then piping it into the headset as your video feed. So it's a cool trick. It didn't bother me, like it actually felt like I was looking through a semi-transparent display, which I just thought was a very impressive technical accomplishment. And it did actually feel like I could kind of control the sort of immersiveness of the experience.
Yeah, so you mentioned these videos they showed us. And I want to say that to me, it was the entertainment-focused stuff that they showed us, so mostly just watching video, that was the most compelling stuff that I got to try during this demo. You know, there's a moment where we started, where I started to watch the Star Wars trailer and you have this like light seal that presses into the device that blocks out all the light around you. And so everything goes black in the exact same way when you're like sitting in a movie theater and right before they start to show the movie, everything goes black. And you sort of have that moment of like, okay, now I'm just going to focus on a movie for two hours. I felt that way in the Vision Pro. And I like that because I don't know about you, Kevin, I am somebody who can like barely watch a movie on my TV without just scrolling on my phone for the entire time.
And I had that sense of relief of like, oh gosh, like maybe if I had one of the things, I actually watch a movie from start to finish.
Wow, yeah, this this piece of technology will definitely save your attention span. That's that always works. Okay, so that was the entertainment piece. Then what I was really curious to see was the productivity piece because Apple is not just billing this as a very cool sort of way to watch movies at home or take these spatial photos and videos. It really wants office workers to buy this and use it for work. And so in a lot of the promotional material, they would show these scenes of people, you know, at their desks and they've got their their Mac monitor in front of them and then they put on their Vision Pro and they can sort of open up windows in space and kind of move them around and resize them, make them as big or small as they want. And basically, people can create their own ideal desk setup and take that with them,
whether they're at a coffee shop or on a plane or sitting at their desk. So I was excited to try this stuff and I want to hear your impressions. My impressions were this stuff is not quite there.
So on the positive side, I would say that this has the best visual fidelity of any headset I've ever used. These are really high end displays. And so at one point, they told me to open Safari, the web browser, visit any website you like. I, of course, entered Platformer to see how that was going to look in mixed reality. It looked great. The the text was very crisp. You know, you could scroll through. I visited some other web pages, like the photos looked really great. And yet I thought like, the amount of ingenuity that has gone into recreating digitally an experience that was already working perfectly fine for me in a laptop feels a little bit crazy here. And so while I appreciate all of the the skill and the creativity that went into making this
thing work, I still couldn't figure out why I would want to do all my web browsing and typing in the Vision Pro as opposed to just like my MacBook.
Well, I've always had the thought while browsing Platformer and reading your articles, like, I wish I could have this just way bigger and closer to my face.
If this could just be on my face, that would be the ideal way to to read my favorite tech newsletter.
A direct feed into your your digital cortex.
So I'm very excited for this feature. But, you know, in seriousness, I thought this was The least impressive part of the demo. You know, I also asked if I could see like something else in the kind of productivity world. Like what else you got for the office worker? And they showed me this part of the demo. They don't think they showed you. That's brag.
Wow.
But it was a feature in keynote. The.
Oh, damn, you got the keynote demo.
I did.
Man.
And so they, this is their like sort of PowerPoint equivalent. And they showed me this thing where in the Vision Pro, you can practice giving a presentation to like an empty conference room or an empty theater.
That's, I would love to see you spending more time just practicing talking. I think it would honestly really improve the podcast.
And I didn't think it was that great a feature. It felt sort of like a giving to me, but it is exactly the kind of thing that office workers are going to try to use to convince their bosses to let them expense a Vision Pro, which I did appreciate.
Well, while you were getting that demo, I was getting another productivity demo, which was Algorithms DJ app. Now, this is an app I've used for many years. I've DJed my friends' weddings using this app.
Wait, really?
Yeah, there is.
I didn't know you were a DJ.
Yeah, I'm not a good one, but like for a close friend, I'll DJ at a wedding. We'll have a great time.
What's your DJ name?
Um, DJ Task Rabbit.
No.
No, yeah, that really is. That's what I.
So.
So, yeah, DJ, DJ Task Rabbit.
My kid's birthday party is coming up. Are you available?
Yeah, but DJ Task Rabbit is very expensive. Just keep that in mind.
It's an easy assignment. All you have to do is play the co-caramel version of Wheels on the Bus 72 times.
Okay, I think I can do that.
Okay, so anyway, you were, you were looking for productivity enhancements for your side gig as a DJ.
Exactly. And so we pull up this app and they, you sort of immediately see a couple of turntables, a couple of records. And I should say, in 2022, Meta released its highest-end headset to date, which is called the Meta Quest Pro. And when I went down to try that, I also got to use a DJ app. And, you know, there was like some cool stuff about it. It was not the DJ app that I, that I used when I was in Cupertino. But one of the things that I don't love about the Meta Quest Pro is that the visual fidelity. Reality is like, okay, but it's not great. It doesn't really feel like you're standing in front of the wheels of steel. Well, in this DJ app on the Vision Pro, the visual quality is so good that you could just kind of pretend that you were DJing.
And once I got the music going, I was, I just intuitively reached down to scratch, you know, do a little bit of wicca, wicca, and it worked perfectly. And like, you know, just as you would want it to. And, and so there was part of it that was like, you know, just pure skeuomorphism of like, we will make a DJ rig and you can DJ on it here in mixed reality. Well, but they also had this sort of enhancement, which I can only describe as like a kind of
3D box that sat on top of the wheels of steel. And you could reach your hand into it. And as you just sort of moved your hand around, you could manipulate the sound. So you sort of felt like a DJ wizard that was just kind of distorting and changing the music that was playing. Did it sound good? No. Would I be asked to leave a wedding if I were using this technology and DJing? Yes, I would. But did I enjoy it during my demo? Kevin, yes, absolutely.
I'm so glad for you. Wow. I'm still reeling from the revelation that you've been holding out your, your secret side career as a DJ.
I love music. Music is very important to me.
Okay, so that's the demo that we got. We also have to talk about what we didn't get to try, which is I wanted to FaceTime call you and see how our personas would look. So personas-
Tell us about these personas.
So personas are this feature, Apple reminds people, it's a beta feature for reasons that will shortly become clear. Basically trying to solve the problem of how do you do a video call using one of these headsets? Because if you, you know, are just doing a normal Zoom call or a FaceTime call, like your face is covered by this headset.
How are you going to look normal on a video call? And Apple's answer to this question is to basically create a deep fake of your face called a persona.
I think I would probably say an avatar over a deep fake.
Well, you know, potato, potato.
But basically, you know, you scan your face and then Apple sort of renders this like 3D model of your face that it then uses as kind of a virtual stand-in for you on video calls. And we were not able to try this, but several other reviewers have been able to, and I will just say, they look very funny.
They do. People, I was reading on threads yesterday, people were saying that your persona sort of makes you look like the PlayStation 3 version of yourself. Where it's like, it's recognizably you, but like, it's a little blocky, you know, it feels like some of the individual polygons are like almost visible. And so, and of course, you're like less expressive in this form than you would be like, you know, using your real face. And so, yeah, people were having a lot of fun yesterday looking at the personas.
Yeah. One person I saw compared them to like NPCs in like, in like a video game who are going to offer you a side quest.
And I just think like, you know, clearly this is a product that's going to get better over time. Like this is the V1. But I think, you know, if someone showed up in a meeting with me as a, as one of these personas, like, the meeting agenda is over. Like, we're talking about that and it's like not going to go unnoticed by anyone in the meeting, at least at first.
There's like, there's, there's so much in the Vision Pro that is like, is sincerely cool. There is some stuff in the Vision Pro that I think might be the best way to do things, particularly consuming video, although even that is kind of still a question mark for me. And then there is stuff that is so much more complicated, so much more expensive, and still obviously worse. And in the social features are really where I think that is the most true. So, again, can all of this get better over time? Yes, I'm sure it will. But the, the current state of the art, this feels a little bit more like a party trick than a way that people are going to be getting worse done.
So, let's do a little summary. What was your favorite thing about the Vision Pro and your least favorite thing?
So, and this is another demo that a lot of people have talked about already, but we did this little dinosaur demo where in the, in the mock beautiful living room, there was a blank wall and as I looked up with My Vision Pro, we started this app and basically imagine it's like, it's like the, the, the wall opens up and all of a sudden you're looking at this prehistoric landscape. And a butterfly flies out and you raise up your finger and the butterfly just flies over and it just sits right on your finger. And you know, you move your finger around, the butterfly moves with you. So that's using the eye tracking capabilities, right? Then a dinosaur.
That's really gonna help you in your career, your side career as a butterfly tender.
Yeah. As a huge lepidopterist, let me just say, I was over the moon. Yes, it's the word.
Wait, what? Lepidopterist?
Lepidopterist. That's the study of butterflies. Jesus Christ.
You know.
Did you go to college? Anyways, so after the butterfly, this dinosaur comes in.
Lepre bob.
Lepidopterist. Do you know Vladimir Nabokov, in addition to being one of the great writers, was also a lepidopterist and named many species of butterflies?
You contain multitudes, my friend.
So then the dinosaur walks into the frame and there's like a little, uh, you know, they're, they're dancing around on the prehistoric rocks or whatever. And then, you know, one of them gets chased away and then the, and I, I don't know that these are velociraptors, but they basically look like velociraptors, would you say?
And you know, one kind of comes up and you can like, sort of, you know, try to pet it and like, maybe it'll let you pet it or maybe it'll roar at you. Why am I talking about this at length? Well, to me, I, I saw this, I thought like, this just feels like new storytelling experiences will be enabled. And I truly did believe after using the Vision Pro, that as the technology gets better, cheaper, more widely adopted, it's getting enabled all sorts of new kinds of storytelling experiences. And you know, may eventually change like the kinds of stories that get told and how they are made. So that was the stuff that I saw that I thought like, oh, there's really something here. And I believe in it more now that I have tried the Vision Pro than I did before I had previously.
And what about your least favorite thing?
My least favorite thing is I don't know if this thing would be comfortable on my face for even an hour and a half. When I have used VR headsets in the past, there was never a day when I wasn't relieved to be taking the headset off of me, where I didn't immediately feel better when I was taking it off. And even though we didn't use the Vision Pro that long, I still did basically have that feeling. I just kind of had like a little bit of tension like at the temples of my head. It was not a headache and it went away very soon after I took this thing off. But it did make me wonder like, if I just want to like, watch, you know, Star Wars on this thing, how am I going to feel at the end of two hours? I don't know the answer to that question. And I gotta say, Kevin, until I know the answer to that question, I don't know if I want to spend $3,500 plus tax.
Yeah. Okay. So what was your favorite thing about the Vision Pro demo?
I loved the spatial photos and videos. I thought, okay, if this were $1,500, maybe that's worth it just for the home movie potential alone. At $3,500, that's a very expensive home movie collection to start building. But I think that's by far the most compelling part of the demo, at least for me.
Yeah. And how about the worst?
And I would say the worst was for me, the productivity stuff. Like, I understand that for some people, if you are a person who works, you know, with a multi-monitor setup and you like having tons of windows all over the place while you work, you are probably the prime candidate for using something like the Vision Pro. But I'm not a person who likes a bunch of clutter and chaos in my visual space when I work. I like focus. I like full screen windows. I like not a lot of other stuff going on. And so for me, it would just be too distracting and that would keep me from using it for work.
All right. So what do we think the prospects of this device, like a year from now, what do you think we will be saying about how the Vision Pro did and like where it goes from here?
I don't know, like, it would not surprise me if this is a fairly, you know, sort of slow launch for Apple. It is a very expensive device. It's a brand-new category. You know, people, it really is one of these things that you need to kind of see for yourself to fully understand. You can't really just like watch a video or read a couple of reviews and get a sense of what it's going to be like to have one of these things strapped to your head. The first version might not be for everyone, but pockets of people will kind of find it and adapt it to their own use cases.
Yeah, that sounds right to me. You know, I think some of the analysts before pre-orders went on sale thought that maybe Apple will sell like half a million of these in the first year. Of course, you know, Apple sells hundreds of millions of iPhones in a year, so this is very small relative to that much, much bigger business. And, you know, on one hand, that's not a lot. On the other hand, that's going to be over a billion dollars in revenue, and that, I think, is going to be enough for Apple to say, "We are going to continue to invest in this." Obviously, Apple has essentially unlimited resources to keep investing in it. And I think we do have enough reason to believe that there will probably be some sort of successor platform to the laptop over time. And, you know, the best way to control the future is to invent it. So I think Apple is going to kind of keep going
in this direction. What I am the most curious about is what creative types do with this? Are there filmmakers and designers that get their hands on one of these and think, "Oh, I could actually make a really cool 10 or 15-minute little miniature story, and I'm going to put that in the Vision Pro App Store and, you know, maybe see if kids like it," and just kind of see where that goes. Maybe in this first year, the install base is just going to be too small for anyone to justify that kind of investment. But I do believe the technology is good enough that if some people took it really seriously and tried, they would make something really cool that people might spend a lot of money to do.
Yeah, I can see that. I also think, for me, the most interesting question about all of this is like, how is this device going to be received, like, out in public? Because if you remember the Google Glass period, where Google had just released this amazing, like, you know, computer that sat on your face and could do all these things...
It was not that amazing.
But they were billing it as this amazing thing. And then they, you know, they sent it out to people and people started showing up in the real world with them. And it was mocked and derided. And one guy even got punched for wearing Google Glass on the streets of San Francisco. And people started calling them glass holes. And it became this sort of social stigma. Like, I don't think that is probably going to happen with the Vision Pro. Cause like, A, it looks a little bit cooler than Google Glass. And B, like, it's not the kind of thing that people are gonna wanna, you know, wear around on the street, uh, walking to and from work all day.
No, and if you see somebody wearing one, you snatch it off their face, "Hey, congrats, you just made $3,500."
But that is like, my question is like, you, sometime in the next few months, you will get on a plane and see someone wearing a Vision Pro. Probably in business class. And I think a very open question, and a very important question is like, how will you feel about that six months from now? Will you be like, "Oh, that's so cool, I want one of those." Or will you be like, "That person is a loser."
Yeah, I mean, I don't know how anybody's gonna feel when they see these things.
Are you gonna buy one?
Not right now. I think, the problem I had with some of these previous headsets that I would try is that they would seem compelling for a week or two, and then I would throw them in a drawer and I'd never turn them on ever again. And I'm not willing to spend almost $4,000 with tax to have that experience. You know, if this thing were like $9.99, I think it becomes, uh, like, yeah, I'll just sort of try that for like, research purposes. At $3,500, $3,800, I don't wanna do that. My caveat is, in two or three months, if some of these experiences that I was talking about start to get made, if maybe people actually find an interesting productivity use, or if people like rediscover a love of movies because it truly turns out to be one of the greatest sort of virtual home theaters around, then I might do it. But this is one where I'm just gonna let other people kind of take the lead here, and
then they can tell me in a couple of months whether this thing is awesome or not. In the meantime, I'm gonna save my money.
I'm, I'm tempted. I gotta say, I gotta say, I'm tempted. Like, I'm not gonna buy one right now because like, I don't have $3,500 just like burning a hole in my pocket. But if I did have like, a distant relative who, who died and unexpectedly left me $3,500, like, I would be very tempted. Cause it is cool.
When we come back, social media went to Congress this week to talk about child safety. We'll tell you what happened.
Support for this podcast comes from Dell Technologies. It's time to put AI to work. Is your business ready? The challenge isn't getting started with AI. It's scaling beyond pilots while maintaining control of your data, governance, infrastructure, and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner. Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.
AI spend is growing fast, but most businesses still can't see exactly where that money is going. Ramp gives businesses visibility and control across AI providers with real-time guardrails by team and model, helping ensure spend goes to the highest ROI areas. Stop guessing where your AI spend is going. Get started with Ramp. And if you're new to Ramp, you can get $250 off at ramp.com/hardfork. Go to ramp.com/hardfork. Ramp.com/hardfork.
Today's AI builders need infrastructure that scales with their ambitions. Irene is a vertically integrated AI cloud platform, delivering data centers, compute, and software for AI training and inference, underpinned by its expansive portfolio of land and grid-connected power in renewable-rich regions globally. Irene's AI Cloud gives direct access to Nvidia GPUs, scalable from hundreds to thousands. For builders, that means direct access to AI compute, rapid scaling, and predictable pricing. Irene, AI Cloud for the next big thing.
Casey, this week, the CEOs of many of the biggest tech companies in Silicon Valley flew to Washington to be dragged before Congress and grilled by a bunch of senators. We've seen a number of these kinds of hearings before, but this was a big one. The CEOs of Meta, TikTok, Snap, Discord, and X all showed up in person in front of the Senate Judiciary Committee for a hearing about harms to children on social media.
Yeah, this has been an issue that has been bubbling up for a couple of years now. More and more states are passing laws intended to improve child safety online. And now Congress is tackling it head on by bringing all those CEOs you mentioned and confronting them with some hard questions.
Yeah, I think you and I have both gotten a little bit jaded about these kind of tech hearings over the past five or six years. We've seen it just time and time again. A tech platform screws up, its executive is hauled before Congress, a bunch of senators or congresspeople sort of pepper them with angry questions that are more like statements. They promise to look at it more and pass some laws, and then nothing happens. The executives go back to Silicon Valley, Congress focuses on other things, and nothing really changes. This one, I think, is potentially very different than that for a couple of reasons. One is, it's about kids. Republicans and Democrats disagree about all manner of tech problems and solutions to those problems.
But I think on the issue of child safety, this is one where there's actually mostly bipartisan agreement that this is a real problem. And I also think it comes at a time where there is actually the potential that something could come out of this. We've seen Congress trying to pass the Kids Online Safety Act, COSA. We talked a little bit about that on the podcast last year. That's a bill that would sort of force social media platforms to be more active in preventing harms to minors. We've also seen a bunch of other laws proposed and states that are taking action to, for example, require social media companies to get parental consent before permitting children under 16 to use their platforms. So Casey, I know that you, like me-
I've seen a number of these hearings come and go without much in the way of action. But did this feel different to you?
Well, I think one way that it felt different was the way that it shook up the cast of characters. Right? So we're sort of used to Congress focusing a lot on Twitter and Facebook in particular. Now we have some newer platforms up and coming. Snap, Discord and X, the former Twitter, all appeared before Congress for the first time. Their CEOs appeared before Congress the first time. And so it showed that Congress is sort of probing at new parts of the ecosystem in an effort to kind of trace how these problems are flowing across platforms.
Yeah, so let's go through some clips from the hearing because it was pretty dramatic. The first one that I think everyone is talking about after this hearing was a moment where Mark Zuckerberg from Meta was asked to apologize to a number of parents in the room whose children had been either victimized or had, in some cases, taken their own lives after being bullied or harassed or otherwise exploited on social media. And he did. He stood up from his chair. He turned around and he directly addressed the parents in the room. Let's roll that clip.
Let me ask you this. There's families of victims here today. Have you apologized to the victims?
Would you like to do so now? Well, they're here. You're on national television. Would you like now to apologize to the victims who have been harmed by your products? Show them the pictures. Would you like to apologize for what you've done to these good people?
I'm sorry for everything that you have gone through. It's terrible. No one should have to go through the things that your families have suffered. And this is why we invest so much and are going to continue doing industry-leading efforts to make sure that no one has to go through the types of things that your families have had to suffer.
If you can't exactly hear him because he is off mic, he's basically apologizing. He's saying, "I'm sorry for the things people experienced." He says, "No one should have to go through what you went through." And then he says, "This is why we're investing so much in trying to prevent these harms to children." Casey, what did you think of this?
So this was a really dramatic and a sort of very rare moment when one of these sort of titans of industry actually has to be in the same room and be confronted by people who feel like they've experienced really harm as a direct result of the software that Mark and his teams have built. So I think that's one that we'll remember for a long time.
Yeah. All right. Next clip. This one was about COSA, the Kids Online Safety Act. And Richard Blumenthal from Connecticut, who was one of the creators of COSA, basically is going down the row of CEOs, one by one, asking them whether they support this bill.
Yes or no, Mr. Citron?
There are parts of the act that we think are great.
No, it's a yes or no question. I'm going to be running out of time. So I'm assuming the answer is no, if you can't answer yes.
We very much think that the national privacy standard will be great.
Mr. Siegel?
Senator, we strongly support the Kids Online Safety Act, and we've already implemented many of its core provisions.
Thank you. I welcome that support, along with Microsoft's support. Mr. Chu? Senator, with some changes, we can support it. No, in its present form, do you support it? Yes or no? We are aware that some groups have raised some concerns. It's important to understand. I'll take that as a no. Ms. Yacarino?
Senator, we support COSA and will continue to make sure that it accelerates and make sure it continues to offer community for teens that are seeking that voice.
Mr. Zuckerberg?
Senator, we support the age-appropriate content standards, but would have some suggestions on how to implement it.
Yes or no, Mr. Zuckerberg? Do you support the Kids Online Safety Act?
Senator, I think these are nuanced things.
That is public, and I'm just asking whether you'll support it or not.
These are nuanced things. I think that the basic spirit is right. I think the basic ideas in it are right, and there are some ideas that I would debate how to best implement them.
Hey, unfortunately...
Okay, so essentially, two of the CEOs in the room, Evan Spiegel of Snap and Linda Iaccarino of X, have broken from the other tech companies and decided to support COSA, this bill that has become extremely controversial in the tech industry. Why do you think they did that?
I think, you know, in Snap's case, they just think it will cost them nothing because a lot of COSA has to do with like algorithmic amplification and stuff that just doesn't really concern Snap as a company that's primarily focused around messaging. With X, and this is not based on reporting, this is just speculation, but I think X probably just wanted kind of an easy win at a time when they're facing some very serious and important questions about why they led deep fake synthetic nudes of Taylor Swift to spread unchecked on the platform until they were getting tens of millions of views just this past weekend. This at least enables them to say, well, look, we did something. We supported your bill, Senator. You know, I have a lot of thoughts about this, but let me just say, wait until Elon Musk finds out what is in COSA
because I suspect he's going to have a strong disagreement with Linda Iaccarino about X's support for that bill.
Okay. The last clip I want to play is from Mark Zuckerberg, who is talking with Amy Klobuchar, the Senator from Minnesota, about why he thinks that the responsibility of verifying users' ages should not fall to Meta and other social platforms, but should instead be handled by Google and Apple, who run the app stores.
I don't think that parents should have to upload an ID or prove that they're the parent of a child in every single app that their children use. I think the right place to do this, and a place where it'd be actually very easy for it to work, is within the app stores themselves, where my understanding is Apple and Google already, or at least Apple, already requires parental consent when a child does a payment with an app. So it should be pretty trivial to pass a law that requires them to make it so that parents have control anytime a child downloads an app and offers consent of that. And the research that we've done shows that the vast majority of parents want that. And I think that that's the type of legislation, in addition to some of the other ideas that
you all have, that would make this a lot easier for parents.
Just to be clear, I remember one mom telling me, with all these things she could maybe do that she can't figure out, it's like a faucet overflowing in a sink and she's out there with a mop while her kids are getting addicted to more and more different apps and being exposed to material. We've got to make this simpler for parents so they can protect their kids, and I just don't think this is going to be the way to do it. I think the answer is what Senator Graham has been talking about, which is opening up the halls of the courtroom so that puts it on you guys to protect these parents and protect these kids, and then also to pass some of these laws that makes it easier for law enforcement.
Okay, so that's the hearing in a nutshell. Congress really wants these tech platforms to do more to protect underage users, and a bunch of senators sort of don't think they're doing enough and want to use new legislation or maybe the courts to go after them for not doing enough for underage users. The tech platforms all say, well, we're doing all these things already, and some of them say we would support legislation like COSA, some of them say, well, we don't think that's the right approach, but we also agree that more needs to be done to protect underage users. Basically, everyone is agreeing that there is an important problem to solve here, there's just some disagreements about how to solve it.
I think it is tricky to talk about. It is very emotional. Everyone wants children to be safe online. I think we have very little agreement at this point about what does safety mean, and I think the fact is that there are just always going to be some risks associated with being on the internet. But if you accept all of that, what is a path forward? Well, the paths, I think, look different depending on what problem you're talking about. But one place to start that I think would actually be productive is talking about this issue of age verification that Zuckerberg brings up in the last clip.
Well, and let's talk about it. So do you think this should be a responsibility of tech platforms to verify how old their users are, or do you agree with Mark Zuckerberg that Apple and Google should take that on?
So I wanted to take a quick step back and say like, we have never been able to mandate age verification in this country because the Supreme Court has repeatedly held that it is a violation of the First Amendment, not because of what it does to kids, but because it places too high of a burden on every adult user of the internet to have to verify their age every time they want to use a website. Okay. So that's why we've never had that in this country. So that brings us to what Zuckerberg says, which is, well, why don't you make Apple and Google do it? And I have to say, I think that this solution is obviously correct, because imagine that your child has come of age where they now have a smartphone and they start downloading apps. And it's not just one app. And it's
not just 10 apps. It's 40 apps. And now, all of a sudden, you are being asked as a busy parent with a job to, in 40 different cases, verify your child's identity. That seems obviously way too onerous for me to pass. We talk so often on this show, Kevin, about the dark side of living in a world where there's only two major smartphone platforms. This is actually a silver lining, because now you have two gatekeepers that can say, you know what, we can actually take it upon ourselves to do the verification. And in fact, as Zuckerberg points out, Apple is already doing this. If you have an Apple Wallet, it's going to ask, hey, are you older than 12? So I want to quickly say, it does not have to be Apple and Google who are doing this. You could also imagine some sort of
industry body that is passing along some sort of token through an API that is made available through iOS and Android, right? So I'm not saying 100% it has to be Apple and Google, but that is the basic level at which this needs to occur, not at the level of the app. What do you think?
I mean, I think I agree with it. I think it's just easier to run age verification through the app stores. You can imagine, you know, you take a new iPhone out of the box, and while you're setting it up, it sort of asks you to verify the age of the person who's going to be using that. And then, you know, that sort of birthdate kind of sticks and is passed through to other applications as you download and register for them too. But I also think like, It's it's good for social platforms to have, if not a an exact number, than than just a a general sense of how young their users are. Right now, you know, a lot of people just lie.
When they register for a new Instagram account or a new Facebook account. If you're underage, you'll just say, you know, I'm 18. And that gets you in. And I think that it's good that tech platforms will soon have to have a better idea of how old the people using their apps are.
Yeah, and you know, we should say like, there is clearly an element of passing the buck in what Zuckerberg says, right? Because the moment that this becomes Apple and Google's responsibility, he basically never has to think about it again. He just flips a switch that says, hey, iOS, if this user is under 13, they don't get to download Instagram and move on with his life. So I can understand why they want this. The thing is, I just think that would probably be a a world where kids are made more safe.
Yeah. This is, as you said, a very emotional topic and one where people, especially parents, have a lot of strong feelings. We did a segment on the show last year about COSA, in which we sort of raised some potential objections to it. And and you especially said that you were opposed to it because it can allow state attorneys general to basically try to crack down on any content that they considered offensive or harmful to children, whether that's pro LGBTQ content or affirming gender content targeted at at kids who may be experiencing some gender dysphoria. You really worried that COSA would have all these unintended consequences. After that, after that episode, we heard from a ton of listeners
who disagreed with us, some of them agreed with us, basically talking about what it's actually like to have kids who use social media platforms, which neither of us frankly does. And I wanted to just bring up two emails that we got in the wake of the COSA segment from listeners and see if any of your views have shifted on this stuff over time. One was from a listener named James, who wrote to us and basically said, "Look, you guys, you talk about this stuff. You analyze this proposed legislation. You do not understand how bad the internet is for kids and how many kids are being hurt online." And this is sort of a point that you'll hear from.
People who are activists on this stuff, they'll say, you know, this is not some small number of teens who are being victimized by horrible things online. This is in fact millions and millions of teenagers. There are studies that have been done by SNAP that found that 2/3 or 65% of teens and young adults in six countries reported that they or their friends had been targeted by some of these online sextortion schemes. There have also been internal studies that have been published from Meta claiming that 1 in 8 Instagram users under age 16 said they had experienced unwanted sexual advances
on the platform in the past week. So, these are the kinds of numbers that I think make parents really start to freak out because all of a sudden, this is not just a few isolated incidents, this is a real epidemic.
Yeah, and those are all terrible things that are worthy of legislation that reduces those problems. But that's not what COSA says. What COSA says is the platforms need to protect minors from harmful content.
Well, who gets to define what harm is? It is going to differ state by state, attorney general by attorney general, political party to political party. And one of the reasons that the founders of this country passed the First Amendment was to prevent these questions from being put into the hands of politicians. So if COSA said, "We want to create a legal framework for ensuring that platforms prevent sextortion," I would be 100% in favor of it. But Kevin, I think anybody who thinks that COSA is going to solve the problems that you're describing is kidding themselves. Because I think in practice, this is going to turn out to be primarily a mechanism for partisan state attorneys general to launch political stunts. I believe Republican AGs are going to sue platforms for showing information
about abortion in states where it is illegal, and then they'll wash their hands of the whole thing and say, "See, we did something to protect the kids. We prevented them from harm."
Well, I just want to clarify that the text of COSA has gone through a bunch of changes in response to some of these concerns, like the one that conservative attorneys general could potentially weaponize it to sort of go after speech they don't like, or speech that is sort of pro LGBTQ plus in some way. And they do seem to be open to further changes. Earlier this month, Senator Blumenthal actually told Politico that he would be open to giving the Federal Trade Commission the sort of enforcement authority when it comes to COSA, instead of putting it in the hands of state AGs. The idea sort of being that since the FTC is federal, maybe it's less prone to kind of politicization than state attorneys
general would be. But I agree that the text of the bill as it is written today does seem to give state AGs some authority to go after tech platforms if they deem them to be sort of not taking enough steps to prevent minor users from encountering harmful content. And I would just say that my views on this stuff have been kind of in flux recently because I agree with you. I think there are problems with COSA as it's currently written. I worry about unanticipated consequences. But we are also starting to learn so much more about the fact that these tech platforms knew they had a problem with minors being exploited, harassed, bullied, and harmed on their platforms, and that they didn't do enough to prevent
it. The Wall Street Journal has been doing a series of really great stories on this stuff specifically related to Instagram and Facebook. They reported on this internal Meta presentation from 2021 that was included in one of these state lawsuits against the company that estimated that 100,000 minors every day receive photos of adult genitalia or other sexually abusive content on Facebook and Instagram. They also describe an internal document that was circulated at Meta, noting that its own recommendation algorithms, in particular this algorithm called "people you may know," which sort of tells you people that you might want to connect with, was known among
Employees to be connecting child users with potential predators. And it also says that Facebook executives were alarmed about certain features in their apps that were harming teens, like these photo filters that mimic the effects of plastic surgery, and that some Meta executives actually wanted to ban these filters, but Mark Zuckerberg refused. So I will say, as we learn more about what was going on inside these platforms over the past few years when it comes to child harms, I am becoming much more sympathetic to the view that these platforms need some regulation to force them to pay more attention to this stuff.
Yeah, I think your points are well taken. It is true that these companies have a lot to answer for. They have contributed to a lot of harm. Often, the voices inside these companies that were raising these issues are drowned out, and it sucks. And I have become much more open over the past several months to the class action lawsuit that was filed by the attorneys general, seeking to hold Meta accountable for everything that you just said. This is just the point where I, as a journalist and somebody who relies on the First Amendment to do my work, I just really don't want to live in a world where the government is writing laws as broadly written, so as to put the government in charge of deciding what speech is harmful. I think that's an extremely dramatic step with a lot of obvious downsides.
And if we wanted to target any of the number of very real problems that you just described, we could take a much more surgical approach.
Yeah. So Casey, where do you think we go from here? We've had this hearing. We now know where the tech companies stand on this one particular piece of proposed legislation, COSA. What do you think happens now?
Well, I think we're going to see a continuation of what we've already been seeing, which is more states passing legislation on this subject. In Congress, we have this perpetual gridlock, but there are many, many states where there is single-party control of the entire legislature. That is why we are seeing so much legislation being passed at the state level, and more and more states are taking a crack at a variety of pretty restrictive things, right? Montana has tried to just ban TikTok in the state altogether. That has been blocked by a federal judge. But I think you're just gonna see state after state, taking swing after swing. All of this stuff is gonna go up to the Supreme Court. And then the question is just going to be, will the Supreme Court actually agree to any of these things that the previous Supreme Courts absolutely dismissed?
This is where my jadedness comes in. I do not think anything is going to happen at the federal level. This stuff has been in gridlock for years. I think that that is going to continue. You know, we're heading into an election. Who knows what's gonna happen in 2025? But I do think you're still gonna see a lot of movement at the state level and in the courtrooms. Because remember, that class action lawsuit against Meta is really starting to pick up some steam. And I do think it is going to cost that company in particular in the end. But what do you think?
Yeah, I think this is gonna be maybe the biggest challenge that social platforms face over the next few years. I think this is an area where the big platforms are really vulnerable. I think legislators and activists and lobbyists know that. And that is why we're going to see them continue to hammer the tech companies on these points.
Yeah, I just wanna say, like, I do worry. I read this great essay by Dana Boyd, who's this researcher who's studied social media for a long time. And she was saying that there is such a risk in the legislation that is now being proposed of falling victim to this idea of techno-solutionism, right? Which is that social networks caused all of our problems with children, and tech companies can solve all of our problems with children. Social networks have a huge role to play. But let's also remember, kids are suffering for a lot of reasons that have nothing to do with who said what to them on Instagram, right? Kids get bullied in schools, actually. Schools have not actually done a lot to solve the bullying problem over the past three decades. So I just want us to keep all of that in mind, too,
because it is so easy to fall victim to the same thing that critics accuse journalists of, falling for the hype that, you know, tech can save the world. If you think that tech can solve all of our problems that tech has created, I also think you're kidding yourself.
Yeah, I think that's right. I think there is a danger of sort of being too deterministic about how the decisions that platforms make affect the experiences that kids have on them. And I actually, like, I don't blame tech companies for having underage users or even for, you know, exposing inadvertently those users to some harms just by virtue of the fact that they're massive platforms that can't keep tabs on everything that's happening on them. What I do fault them for and what I do think a lot of parents are going to fault them for is that after they learned that teenagers and young people were having bad experiences on their platforms, they just didn't do enough to stop it.
And in some cases, they rejected proposals that would have helped kids on their platforms have a safer experience because they cost too much money, or they involve too much bureaucracy or oversight, or they just worried that calling attention to the fact that there were underage users on their platforms would open them up to new forms of scrutiny. So I think this was a huge mistake tactically that a lot of these platforms made in not talking about this sooner. And I think those days are over because now parents are pissed off and they want answers.
Well, Kevin, what would you say if I told you that there was another company that knew about a huge problem and when confronted with it did not do the right thing?
Come on!
That's right. Coming up after the break, we'll tell you how a car accident took down Cruise.
Started with AI, it's scaling beyond pilots while maintaining control of your data, governance, infrastructure and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner. Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.
AI spend is growing fast, but most businesses still can't see exactly where that money is going. Ramp gives businesses visibility and control across AI providers with real time guardrails by team and model, helping ensure spend goes to the highest ROI areas. Stop guessing where your AI spend is going. Get started with Ramp. And if you're new to Ramp, you can get $250 off at ramp.com/hardfork. Go to ramp.com/hardfork. Ramp.com/hardfork.
Ripple has been powering modern finance for a decade. It's not our first trip around the blockchain and it won't be our last. We're leading the charge to bridge the gap between existing financial systems and the future of finance. You can have it all, legacy, stability and digital innovation, systems that work with your other systems. Because with the right partner, the question isn't crypto or traditional finance. The question is simply, why choose? A new standard of modern finance. It's happening with Ripple. For informational purposes only, availability of products can vary. Maybe provided by different Ripple entities.
Hey Kevin, remember that one time we interviewed the CEO of Cruise?
Yes, the self-driving car company.
Yeah, nice guy. Whatever happened to that?
Well, it's interesting you ask. It's been a very rocky few months for the Cruise company. And Kyle Vogt, the CEO, who we interviewed on the show last year, stepped down in November, following a big scandal involving regulators and a serious accident and a cruise vehicle called Panini. And it all culminated recently with this big report that was prepared about that incident and kind of what has led Cruise to the place it is now. So. As of today, the company has lost its license to operate driverless vehicles in California. It has suspended its operations across the country. The company's basically entire leadership has been replaced. A quarter of the staff of Cruise has been cut.
It's being investigated by the DOJ and the SEC. And now, as of this week, GM announced that they are cutting their investment in Cruise by half, about a billion dollars this year.
So what I'm hearing is that this is a story about how one car accident destroyed an entire company.
It's sort of about that, but I think it's also about kind of self-driving technology as a whole and some of the trade-offs that we're going to see as these cars become more widespread. And I really think we should talk about the Cruise incident today because it's not only just kind of like a juicy story, but I think it also really symbolizes the issues with this kind of move-fast-and-break-things mentality that a lot of tech companies have, even in areas like self-driving where the costs of a mistake are so grave.
Yes, and I should say, I have read the report that came out of this and it is, so there's some really disturbing stuff in there, and I think it is worth going through it.
Yep, and it really matters what happens to Cruise. This was one of only two companies that was offering self-driving sort of halable rides in San Francisco and other cities. The other one, Waymo, is still operating. You can take their cars today in San Francisco. But Cruise was a big player and a company that GM in particular had bet a lot of money on sort of being the first to bring this technology to market. And I'm a person who thinks that driverless cars are one of the most important technologies out there today. I think that roads are tremendously unsafe and self-driving cars, if they work well, they could make things a lot safer, but there are also still some issues with them.
And so I think it's worth taking sort of a closer look at what happened at Cruise because I think some of these same dynamics could play out across the industry in the coming years.
Yeah. So let's get into it.
So I think to tell this story, we have to rewind the clock to October 2 of last year. And that night in San Francisco, there was an accident. Now, this accident was not caused by a cruise self-driving car. It was caused by a human driver who basically did a hit and run on a pedestrian in downtown San Francisco. That pedestrian was flung by this human driver collision into the path of a cruise autonomous vehicle. The cruise vehicle tried to slam on the brakes, but it ended up hitting her. And then it tried to execute a pull over maneuver, which is what it's programmed to do when it.
And when you say execute a pull over maneuver, you mean pull over.
Yeah, I'm just using the language that Cruz used in this report. So it tries to pull over because that's what it's supposed to do. And in the process of pulling over, it drags this poor woman about 20 feet to the side of the road while pulling over.
Because critically, and here is where the AI messed up. The AI thought that the woman had hit the car on its side and so that it would be safe to pull over. In fact, the woman was in the front of the car. And so as the car pulled over, it dragged the woman 20 feet and she was left in critical condition.
Yes. So very sad story. But what happened next is that in responding to this incident, Cruz was supposed to meet with a bunch of regulators to basically debrief the incident and see what could be done to fix this kind of thing in the future. And in the process of those meetings with government officials and regulators, Cruz executives basically left out the fact that their car, while it had not caused the initial injury, did drag this woman about 20 feet before coming to a stop.
So this winds up becoming the key omission that winds up, you know, essentially causing the company to collapse. In the immediate aftermath of the accident, though. Cruise really only wanted to communicate one message, which was, hey, we didn't cause this accident.
Right.
So this is the narrative that they take out to the media. This is the drum that they beat. And they're so fixated on the fact that they did not cause the initial accident that they wind up making a bunch more mistakes.
Right. So the basic story is, you know, there's this accident. Cruise responds, meets with all these regulators and government officials to talk about what happened, plays them footage from the cameras that were on the car at the time of the incident, but kind of glosses over the fact that their car dragged the woman about 20 feet before coming to a stop. They get caught sort of doing this kind of selective presentation. And as a result, GM, which is the majority owner of Cruise, sort of replaces the senior leadership of the company and commissions a law firm to basically investigate what happened and come up with a report. And this report was made public just recently. And it is a wild document. Did you read it?
I did read it, and it was probably the most interesting thing I've ever read that was written by Quinn Emanuel Erkhart and Sullivan, the law firm that did the report.
Yeah, I love these kind of lawyers' reports because it's like the precision in these documents, the timestamps on all the Slack messages. It is just like, you do not want this kind of thing to happen to you.
You don't. And yet, at the end of the day, Cruise is their client, and they do pay for everything that's happening. And so I do feel like the law firm sort of writes about it in a hilarious, letting them off the hook way that I would like to talk about.
Yeah, so let's talk about this report.
Okay.
So the report basically says that, you know, in the immediate aftermath of this incident, Cruise was sort of worried because initial media reports suggested, some of them, that Cruise's self-driving car had, in fact, caused this accident. That this was a case of a driverless car hitting a pedestrian, which would be a big story.
Yes. And it is the sort of the story that the entire press corps has been waiting for ever since these self-driving cars got on the road. When would one of them cause a potentially fatal accident? And so now it happens. And so Cruise, which I'm sure already had a full plan, ready to execute the moment that this happened, they spring into action.
Right. So on October 3rd, after this incident happens, Cruise executives and employees are kind of trying to piece together what happened. They're looking at the footage from the cameras on these vehicles. And at around 3:45 a.m., according to this report, a Cruise employee first sees the full video of the incident and learns that in addition to kind of hitting this pedestrian and stopping, their car also pulled over and dragged the woman with it. So the company's senior leadership, including Kyle Vogt, meets to discuss this new video and tries to decide whether or not they're going to sort of update the media with this new detail. And they decide not to.
Oh, that seems like a mistake in retrospect. Yeah.
Yeah. So they also meet with the mayor of San Francisco's transportation advisor. They don't mention the dragging part of the incident then. They meet with a bunch of regulators, including NHTSA, the DMV, and the California Highway Patrol. And same thing, they talk about this incident. They show this video of the footage from some of the cameras, but they do not sort of proactively bring up the fact that their car dragged this woman after hitting her.
And here is how this is described by Quinn Emanuel, by the way. Quote, "In each of those meetings, Cruise had the intent to affirmatively disclose those material facts by playing the full video and letting, quote, 'the video speak for itself.' Because Cruise adopted that approach, it did not verbally point out these facts." So I just like, that is such a hilarious word salad to explain why after you drag a woman 20 feet and you're meeting with regulators, you do not mention that you dragged a woman 20 feet.
But you're leaving out the best part of this, which was that apparently Cruise tried to play video of the crash to officials. But the person who was playing it, their wifi was not up to the task. They had bandwidth issues and a poor internet connection that prevented the regulators from seeing the complete and clear full video of the accident. And not only did this happen in one meeting, it apparently happened in three different meetings. And I would just like to say to this Cruise employee, Please call Comcast. Upgrade your plan. When you are showing a crash video to regulators, you're definitely going to want the gigabit fiber.
This truly is the most dog ate my homework excuse for not being able to show this video to regulators that I can even imagine. And when I saw it, I really did gag a bit.
Unbelievable. So the entire future of the driverless car industry, we now know, may hinge on one cruise employee's shitty wifi.
Wait, I just want to share Quinn Emanuel's conclusion about this, which is, quote, "Even after obtaining the full video, Cruise did not correct the public narrative, but continued instead to share incomplete facts and video about the accident with the media and the public. This conduct has caused both regulators and the media to accuse Cruise of misleading them." It's like, yeah, you think?
Yeah, so there's a lot more in this report. It's very detailed. It's many, many pages long. And, you know, a lot of it is just kind of internal fact finding. But basically, the overarching conclusion that this investigation draws is that Cruise executives and leadership materially misrepresented what happened on this October 2 night with this pedestrian to regulators, which is a very bad thing to do if you are a company in a heavily regulated industry like transportation.
So, you know, the law firm that did the report, I think disagrees with the idea that this misleading was intentional. The report says, quote, "Despite the failure to discuss the pullover maneuver or pedestrian dragging with regulators, the evidence review to date does not establish that Cruise leadership or employees sought to intentionally mislead or hide from regulators the details of the October 2 accident. Instead, they attempted to show the full video of the accident in good faith, but with varying degrees of success due to technical issues." And so I would not say this rises to the level where you can truly call it a cover up, but it does sort of feel like a cover up by omission, right? Where like, the company did go out of its way to just not say what had happened. And when the inevitable happened, which was that regulators finally see the full video,
it finally becomes clear to them what has truly happened. Well, of course, they wind up being way more upset than they would have than if Cruise had just been honest with them from the beginning.
Yeah, and I've talked to some people who were involved in this situation over the past few weeks and, you know, the folks on the kind of GM and regulator side of this just basically paint Cruise as this kind of reckless startup that was more interested in kind of scaling their business and getting as many of their cars onto the road as they can and sort of taking market share away from Waymo than they were about making sure that all their vehicles were safe. And people on the other side, the sort of more cruise-friendly part of this say, "Well, we have our risk calculations all wrong when it comes to driverless cars. You know, we shouldn't be asking, are these things perfectly safe? Are they never going to be involved in an accident? Are they never going to hurt someone? We should
be comparing them against human drivers who we know cause accidents and hurt people all the time." And so in their minds, this is just kind of like an overly cautious set of regulators and overly cautious corporate parent kind of looking over this with just the wrong set of risk calculations in their brains.
Well, I have to say, I think that that really misses the point. I think that the real story here is even sadder because, look, Kevin, the moment you decide to create a self-driving car company, you have to prepare for the extreme likelihood that at some point there is going to be a serious injury accident and even a fatal accident, right? And I'm sure Cruise had done a lot of thinking leading up to this moment about how it was going to handle that situation. Again, human-driven cars are killing people every day. This should not surprise anyone. Our hope is that these things become safer. And what this incident is really about is that the AI made a mistake and probably a mistake that a human being would not make. And that is tragic and that is worth exploring. But that's not why Cruise
barely exists anymore. It barely exists anymore because in the aftermath of this, the company could not just say like, "Oh, we made a mistake. Let's own up to our mistake and let's fix it and move on." If they had done that, I truly think Cruise vehicles would still be on the road today.
Yeah, that's a good point. And we should say like Cruise, it still exists. They're still planning to expand their services in the future.
I mean, they say they are, but like, what happens? Like, is Cruise going to continue to exist in any meaningful way, do you think?
Look, I think GM has sunk a lot of money into this company. These driverless taxi services, they are just hemorrhaging cash. It's very expensive to build and maintain and keep these cars on the roads. And they're not really making money from them right now. And so I think GM probably sees this as an opportunity to kind of cut back some of its losses. But look, I do think driverless cars are not going away. The technology is here. So what do you think happens now to the dream of driverless cars? Like, does this push the timeframe out by several years? Because now we only have one driverless car company offering rides instead of two?
Or like, what do you think happens to Waymo, for example, as a result of this?
I mean, I have to believe that it does slow the progress of this industry. I think that competition makes industries grow faster. And now that, as you point out, Waymo is the only self-driving car company on the streets, that progress probably is going to slow a bit. So that's kind of what I'm expecting to see. I'm very curious to see where Cruise is a year from now. Like, on one hand, I agree, GM has invested so much money, you can't really imagine them pulling the plug. But on the other hand, every action that they've taken since this happened has given me the impression that they really are not thrilled with this crew. So we'll have to see, but it's just a terrible black eye for the industry. And I do worry it could cost people's lives. What do you think?
So I think this is going to slow the self-driving car industry, less on the consumer adoption side than on the regulatory side. We're already starting to see evidence that this is expanding just beyond Cruise. Last week, the city of San Francisco sued a state commission for allowing Waymo and Cruise to expand to the city. So this is going to result in collateral damage to Waymo too, even though that company appears to have done nothing wrong. And in fact, their safety record is quite good.
So Waymo could actually go away, even in San Francisco now?
You know, that remains to be seen. I think there would be people who would be upset if that happened. And I also think there'd be people who would probably be pretty happy about it.
Now, we always try to leave people with a bit of hope, or at least something, you know, practical that they can do. So here's what I'm gonna say. The next time that there is one of these self-driving car accidents, and investigators are trying to get to the bottom of it, and the person trying to show you the video says, "I'm sorry, my internet just isn't working," say, "You know what? We're just gonna pause the meeting. Why don't you go to a coffee shop and just put the video in a Dropbox folder, okay? And then we'll download it, and then we'll look at the video. And then you can tell us what you think happened in the video, and then we can sort of, you know, bypass this whole issue that seemed to plague us this time.
Support for this podcast comes from Dell Technologies. It's time to put AI to work. Is your business ready? The challenge isn't getting started with AI. It's scaling beyond pilots, while maintaining control of your data, governance, infrastructure, and costs. Dell AI Factory provides the integrated foundation you need to scale AI securely and efficiently with one trusted partner. Dell, AI on your terms. From idea to impact. Learn more at dell.com/yourwaytoAI. That's dell.com/yourwaytoAI.
Ripple has been powering modern finance for a decade. It's not our first trip around the blockchain, and it won't be our last. We're leading the charge to bridge the gap between existing financial systems and the future of finance. You can have it all: legacy, stability, and digital innovation. Systems that work with your other systems. Because with the right partner, the question isn't crypto or traditional finance. The question is simply, why choose? A new standard of modern finance. It's happening with Ripple. For informational purposes only. Availability of products can vary and may be provided by different Ripple entities.
Hard Fork is produced by Davis Land and Rachel Cohn. This episode was edited by Paula Schuman. Today's show was engineered by Chris Wood. Original music by Rowan Nemezto and Dan Powell. Our audience editor is Nell Galopoli. Video production by Ryan Manning and Dylan Bergerson. If you haven't already, check us out on YouTube at youtube.com/hardfork. Special thanks to Puywink Tam, Kate LaPresty, and Jeffrey Miranda. You can email us at [email protected] with your full Vision Pro review.
You can't undo everything in life, but you can undo prediabetes. More than 2 in 5 adults in the U.S. have prediabetes. Most don't know they have it. But with early diagnosis and action, you can delay or even prevent type 2 diabetes. Visit doihaveprediabetes.org to take the 1-minute risk test and start undoing the things you can. Brought to you by the Ad Council and the Centers for Disease Control and Prevention.
Page updated October 12, 2026 · transcript and summary are machine-generated
