

GTA VIは人間の手による金字塔だ
Hard Fork
要約
Hard ForkがGTA VIの発売を取り上げ、Times記者Joe Bernsteinが3週間のプレイ体験を語った。13年ぶりの新作の規模、風刺の難しさ、AI時代における「手作り」の価値が議論された。後半はNVIDIAのJensen Huang復帰論、AIが発見した脆弱性へのパッチによるとされるネットワーク障害、Anthropicの上場(時価総額2兆ドル超)を扱った。
この回のポイント
- ●GTA VIは約5,000人のRockstar社員が13年かけて作ったとされ、マイアミ周辺を細部まで再現した巨大オープンワールドで、Rockstarは制作にAIを使っていないと説明している。
- ●Bernsteinは、買い物や筋トレ、犬の飼育など日常的な要素の作り込みが印象的だと語り、各要素が相互に強化し合う設計が優れているとした。
- ●テックの偏在する社会を風刺する難しさに対し、ゲーム内に配車アプリや強盗といった要素を組み込む手法が巧みだと評された。
- ●「Claudeで数時間で作れるゲームもある」という問いに、Bernsteinは重要なのはセンスだと答え、将来については「1年先より先は分からない」とした。
- ●NVIDIAは上場企業で取締役会に実権があるため、Jensen Huangの復帰はOpenAIのSam Altman復帰劇とは事情が異なるとGriffithは述べた。
- ●相次ぐ大規模障害について、Warner上院議員はAIが見つけた脆弱性への急ぎの更新が原因だと説明していると紹介された。Anthropic上場は通常のIPOのように語られることに違和感が示された。
章立て
GTA VIは現代の大聖堂
13年ぶりの新作の発売と、AIを使わず手作業で作られたという位置づけを紹介する。
規模とディテールへの執念
Bernsteinが、巨大な世界や日常的な動作の作り込み、制作の長期化の理由を語る。
GTAの歴史と情報統制
出演者の初期作の思い出、3D化の衝撃、論争、Rockstarの厳しい秘密保持を振り返る。
風刺とテックの描写
風刺が難しい時代に、ゲーム内のUberやクリプト、Amazon風企業などをどう扱うかを議論する。
最後の大型ゲームとAI時代
大作の希少性、恋愛要素、AIで作れるゲームとの比較、職人性について語られる。
Jensen Huang復帰論
NVIDIAのガバナンス、復帰を求める声、取締役会の姿勢と業績を論じる。
相次ぐネットワーク障害
航空や金融などの障害と、AI発見の脆弱性への更新が原因とする説明、陰謀論への見方を扱う。
Anthropic上場
時価総額2兆ドル超の上場の意味、OpenAIへの影響、安全性の語りが通常のIPOと混在する違和感を議論する。
解説記事
AI・テック番組Hard Forkのこの回は、まず異例の導入で始まる。Max Reedが、今週のテックニュースより先にGTA VI(Grand Theft Auto VI)を語りたいと切り出し、Times記者Joe Bernsteinを迎えて3週間のプレイ体験を聞いた。後半はNVIDIAのJensen Huang復帰論、相次ぐネットワーク障害、Anthropicの上場を、Erin GriffithとMike Isaacらが議論している。
「大聖堂」としてのGTA VI
Reedは、GTA VIを中世の大聖堂にたとえる。13年の歳月、ほぼ5,000人のRockstar社員、212GBのデータで築かれた、マイアミの巨大なデジタル模型だという。Rockstarは制作にAIを使っていないと説明しており、番組は「この規模でこの手法で作られる最後のゲームかもしれない」という見方を示した。
Bernsteinによれば、舞台はマイアミ市街だけでなく周辺の郡まで広がり、60種類の鳥がいて、犬を飼うこともできる。印象的なのは派手な要素より、ジムでの多彩なトレーニングや服装といった日常の作り込みだと語る。ゲームの各システムが互いに強化し合う設計も、評価点として挙げた。制作が長引いた理由については、作業環境の問題への対応や、ディテールに終わりがないことを挙げている。バルコニーにもたれた腕の肉が自然に広がる描写も例に出された。
風刺とテックの描写
GTAは本来、社会を風刺するシリーズだ。しかし現実が風刺しにくい時代になったことが議論になった。Bernsteinは、序盤の主人公が「無能で貧しい犯罪者」で、車を盗むよりバーガーキングを襲って配車アプリで移動する方が楽だという皮肉な設定を紹介する。テックに覆われた生活は、犯罪の逃避空想にさえ入り込んでいるという解釈だ。ゲーム内にはAmazonに相当する企業やクリプト関係者も登場する。なお、制作開始から発売までのリードタイムが長いため、時事風刺は難しく、COVIDは扱われていないという。
なぜ今これほど注目されるのか
Isaacは、予算の大きい大作ゲームはもはや稀で、これが最後に近いと語る。無料のFortniteなどが主流になったことも背景に挙げた。Bernsteinは、RockstarにはGTA Onlineという収益源があるため新作を出す必要は薄かったとし、制作者の作品への愛着が動機だと見ている。AIの話題では、Claudeなどで短時間にゲームが作れる状況と比べて、Rockstarには優れたセンスと膨大な時間・資金があると主張した。ただし、AIがどこまで追いつくかは「1年先より先は分からない」としている。親会社Take-TwoのCEOがAI不使用を宣伝している点も紹介され、手作りの質感がセールスポイントになっていることがうかがえる。
NVIDIAの騒動とガバナンス
後半では、辞任したJensen Huangが復帰の意向を示した件が取り上げられた。Griffithは、Sam AltmanのOpenAI復帰劇との違いを整理する。NVIDIAは上場企業であり、取締役会が独断で決めれば株主訴訟のリスクがある。また創業者支配型の株式構造を持つ企業が多い中、NVIDIAには実質的なガバナンスがあるという。復帰を求める声にはMusk氏らが含まれると紹介されたが、取締役会長は復帰させない意向を表明している。四半期売上が初めて1,000億ドルを超えたことも報じられ、短期的な業績は堅調だと語られた。
相次ぐ障害とAnthropic上場
過去10日ほどで、Outlookの停止、FAAによる全米の航空機の運航停止、United便の停止、Apple Payの障害などが相次いだ。番組によれば、Warner上院議員はAIが発見した脆弱性への更新作業が原因だと説明している。一方、Outlookの不具合は人為的な原因で、AIは修正を助けたとも紹介された。出演者は、AI企業が長年インフラへの危険を語ってきた以上、人々が真っ先にAIを疑うのは自然だと述べた。
Anthropicは上場初日、1株111.50ドル、完全希薄化後で2兆ドル超の評価額で取引を始めたとされる。Griffithは、AIバブル懸念やNVIDIAのスキャンダルの中でも資金が集まったことに驚きを示した。公開市場からは2,000億ドル超を調達したという。一方で、議論は売上や信用格付けといった通常の企業の話題が中心で、安全性や再帰的自己改善への懸念との落差が「不思議な感覚」だとも語られた。
まとめ
今回の回は、手作業の積み重ねが生む価値と、AIが産業や社会基盤を揺さぶる現実を対比的に見せている。日本の読者にとっては、クリエイティブ産業でAI不使用が訴求点になりつつあること、AI発見の脆弱性への対応が障害リスクとなりうること、AI企業の上場が一般企業の指標で語られ始めたことが注目点だ。いずれも発言者の見解や報道に基づく部分が多く、今後の検証が必要である。
文字起こし(英語・自動生成)
Grand Theft Auto 6 is a cathedral. Look, I know it's been a busy week in tech news. Anthropic just went public. Jensen Huang wants to return to NVIDIA. The technological infrastructure our entire country runs on seems to be crumbling in front of our eyes. I promise we'll get to all that stuff in this episode. But first, I want to talk about GTA. Grand Theft Auto 6 was released on Thursday to rapturous reviews and about $1.5 billion in pre-sales. It's being called a masterpiece and an astonishing American epic. Grown adults around the world stayed up till midnight to download it and cleared their schedules to play, all but weeping with praise over a game that's fairly described as a crime simulator with a South Park sense of humor. And you know what? I get it. It's been 13 years since the last Grand Theft Auto game. We've had three presidents in that time. There was a pandemic. They need computers that can talk to you. Throughout that time, nearly 5,000 people have been toiling at Rockstar Studios. Those countless hours of labor have now crystallized into 212 gigabytes of code downloaded to PlayStations and Xboxes across the globe over the past week that describe an enormous digital simulacrum of Miami.
If a large language model is a statistical model of the world, GTA 6 is a qualitative one. You can still steal cars, rob old ladies, and jump off buildings. But you can also adopt dogs, put up drywall, fall in love, buy your partner flowers, hold their hand. It's mesmerizing, strange, funny, and astonishing. All of this has been done, Rockstar says, without any AI assistance. This is what I mean when I call it a medieval cathedral. A generational project, built by hand, stone by stone, asset by asset, rocket launcher by rocket launcher. I can't believe I'm saying this about an entertainment product with an in-game porn site called Booty Bang, but this is a monumental human achievement. And given the direction the video game industry and really all entertainment is headed, that probably makes GTA VI the last game of this scale made this way. Times reporter Joe Bernstein has had access to the game for three weeks now. I wanted to ask him, how did a game like this get made? Why does it have such a hold on our imaginations?
And can I come over to play it on his PS5? I'm Max Reed. This is Hard Fork. Welcome to Hard Fork. I am here, as usual, with Times Tech reporter Erin Griffith. Hello. And today we're joined by journalist Mike Isaac, who is soon to start at The Atlantic but is currently extremely unemployed. Hi, Mike. Hello. And we're extremely excited to welcome Times reporter Joe Bernstein. Joe has spent the past several weeks living out a fantasy shared by millions of people around the world. He is one of a small handful of people with early access to the latest game in the Grand Theft Auto series. He's written a really smart and funny piece about it for the Times Magazine. And I want to talk about this both because I am one of these insanely jealous people, but also because this game is a huge deal. It is, frankly, a really monumental achievement. It's been 13 years since the last game, which was a huge iconic blockbuster that people still play all the time. Some huge number of people helped to make it.
It cost a small nation's GDP or so we suppose. Joe, since you're one of the few people who has gotten to spend time with this game, and you also went to visit the headquarters of Rockstar, the company that makes GTA in Edinburgh, could you tell me a little bit about the scale here, both in terms of the resources that went into it, but also the actual game itself? Sure, yeah. I mean, sometimes we think of video games as having, like, annual editions, like the new Madden or the new FIFA, the new Call of Duty. This is something rare. I mean, three presidential administrations since the last one, right, which came out in 2013. These Grand Theft Auto games are giant open worlds, meaning it's like a huge virtual environment that will be modeled after an American metro region. So the first one took place in Liberty City, which is New York City. Some take place in Vice City, which is Miami, and the new one takes place in Vice City. Several take place in a sort of satirical version of California. And the game they've made is astoundingly huge. So Rockstar, and a lot of this reporting comes from a great Bloomberg Businessweek piece by the games reporter Jason Schreier.
Rockstar staffed up, like doubled in size, something like that, to nearly 5,000 employees to make this game. And they're working on a scale that's enormous. This is a huge open world that sort of encapsulates not just Miami, not just South Beach, but all of Metro Miami, and then sort of like the South Florida of the mind. So, you know, everything from Parrothead, Keys Life, to like Florida Project, like greasy motels in OxyContin. It's just a really all-encompassing thing. Mike is nodding. Sounds like my unemployment status. So it's huge. It's like physically huge. You can run from one end of Miami to the other, and not just Miami. You can run from one end of Miami-Dade County to the other. But it's also like every single step along the way, there's a shop you can enter. There's a car you can steal. There's a dog you can pet. There's 60 different species of bird in this game. Right, and not only that, like, you're free to ignore a lot of that. Phil agrees. So, like, yeah, there's birds all over the place, and I haven't, like, you just told me before we started the show that you can own a dog in the game.
And I didn't even know that, and I've been playing it for, like, 50 hours. I should say, not only can you own a dog in this game, this came up in the context of seeing a clip where a dog was riding a jet ski with the game's main character sitting behind him. So not just that you can own a dog in the game, the dog can be your friend. The dog can drive a jet ski. Yeah, I'm just hearing this now for the first time. So, wait, sorry. Can you say you've been playing it for 15 or 50 hours? Probably close to 5-0. Wow. Yeah, yeah. Those are rookie numbers, Aaron. And that's like, you're saying, wow, I'm going to lose my entire life on this game. Yeah, I mean, Gene Park, who is the Washington Post game critic, says he's already spent 170 hours playing the thing. So I'm 50 hours into it, and according to the game's sort of, like, progress meter, I've completed 80% of the main story. And, you know, just being completely, like, out of the loop on this kind of stuff, Is there a maximum or is it just infinity that you could play hours of this game? Like for 13 years until the next one comes out?
So, yeah, there's like a long answer and a short answer. The short answer is you could do everything in the game. The long answer is that at some point Rockstar is going to make an online version of this. And in the online version, there's a lot of user-generated content. And so that, I mean, people are still playing the online version of the fifth game, which did come out 13 years ago. So it is this kind of like almost the closest thing we have to like a simulation of the real world. And part of the attraction, and I imagine we'll spend a lot of time during this conversation talking about why this is so cool and such a big deal to so many men who have like real responsibilities in their lives. Part of the attraction is that you can kind of do anything you want. You know, you can have dogs riding jet skis. There's a lot of crime you can do, but there's all kinds of other things you can do. What are some of the like crazy things that you've seen or done or tried to accomplish while you've been playing? Yeah, I mean, it's funny because it's a video game. The more quotidian the thing you do, the crazier it seems. So you spend a lot of time at first if you want to and later because you have to for plot reasons going to the gym.
And just the sheer number of exercises you can do in the gym is crazy. It's like deadlifting, squatting, pull-ups, bench press. And then the number of outfits you could wear to the gym. It's funny talking to you about it because you keep talking about things you can do in the game. And it's not you're never saying, well, you can get a laser gun and you can take off in a spaceship. You're saying things that we do in real life all the time, like wear clothes and lift weights. And I think that there's like something to be said about like how much time was spent, like really going into into making all these sort of quotidian aspects feel realistic and pliable and like concrete for people who are actually gaming. And fun. I mean. it depends on your it's a famously hard it's a famously hard word to define what I would say is that when the game is really working the kind of the systems which is how game developers talk about the things you can do in the game
are sort of all mutually reinforcing so like the weight lifting yeah it's like pressing buttons but then the more you lift weights like you're better in a physical fight or you can run for longer or jump. I don't know if you could jump. Again, you're still just describing real life. It's very funny to be hearing you describing the mechanics of a game and you're just describing the way the world works. Yeah, and then there's all the things you do that are not sort of what Aaron or Mike or you Max would do on a daily basis. So there's bank robberies and there's, you know, jet skiing with an AR-15 and, you know, all that kind of good stuff. Parachuting out of a boat plane. Hard Force did not sanction jet skiing with an AR-15, only with a dog. But part of what I'm asking is, is the reason this game took 13 years to make that it required this level of sort of detail and freedom for the player to play? So part of it is that the company made this huge game in 2018, and they focused all of their resources on this game, Red Dead Redemption 2.
Jason Schreier's piece was very useful because it got into some of the kind of internal rock star political reasons. The game took a long time. They had some labor issues that they sort of had to address workplace culture. And this, yes, this sort of sheer level of detail. I mean, this got cut from my piece, but like when I was in Edinburgh playing the game, or rather watching one of the executive producers play the game, they leaned the main character on a balcony and the flesh on his arm splays out naturally on the sort of stucco of the balcony. And it's the kind of thing that, like, you know, obviously happens in real life, but, like, if it didn't happen in a video game, you wouldn't notice that it didn't happen. You also wouldn't necessarily notice that it did happen. It's just the detail in the background that sort of sucks you further into the game. And the sort of, like, the math and the time involved in getting that right is pretty staggering. And then when you see that applied across the board, it becomes clear why the game took so long.
And one of the things these guys, these rock star executives were talking about was also like when you're making something that is so fussy and detailed, there's really no end to it. Right. Because there's nothing more detailed than the world. And so you understand like why, like why it took them so long and how obsessed they got. And that's actually really cool. Like it's it's fun to experience someone else's obsession in that way. You're reminding me of this old. So I'm going to go into like nerdy dreamcast, like old console guy. And there was a game that came out years ago called Shinmu, which is like this Japanese game. Yeah, yeah. And like everyone was marveling at it because you could pick up a cup or a Coke can or whatever. And half the people had Aaron's look on their face, which is like, who cares? Like, why do you want to like pick up a coffee cup or whatever? And the other half, I think, really appreciated the amount of work and labor that went behind making an environment that was as close to real life as possible. And these are things that gamers don't really get all the time in the vast majority of games out there because you can only, it takes a lot of work to go into creating like a virtual world to the degree that these games are.
Okay, just like in the spirit of open inquiry to understanding what is such a, what makes this such a cultural phenomenon is, can you guys maybe talk about like when you first, you know, encountered Grand Theft Auto in your lives and, you know, why it's such a big deal? I have vivid memory of this, if you will allow me. Please. I was a young boy in Texas walking into the electronics boutique back in the day where you could still buy huge boxes, physical boxes, instead of downloading these games. And it was the first one. It was what Joe was talking about. It's top-down. It's very, very, like, rudimentary if you're looking at it by today's standards. But I would argue it was of this era where these type of games were also pretty radical in a way. Like, you could, like, run into people on the street. There were, like, congressional hearings about these types of games, right? There were people that got very mad at this stuff.
But as, like... Just because it was, like, violent or, like... Yeah, violent. And, like, I mean, I was 13 and still bought the game, right? So, like, there was all this stuff around it. But I do think that this was... It was formative also in the sort of open world aspects. A lot of the, like, instead of a linear game, you can run around and explore things. And that was, I don't know, like as a young, as a young nerdy guy who likes looking at these things, it was a little more expansive, let's say. Yeah, I think I got into it like around the same time as Mike. And I remember renting, I didn't have a console. And I remember renting the GTA 2, which was another one that was sort of top down and very rudimentary. And it's a very it's like a it's a it's a development of the game from that kind of trollish like juvenile. It's still quite trollish and juvenile, but it's like remarkably like trollish and juvenile within the context of this remarkably sophisticated engine for recreating the world is is like sort of difficult to wrap your head around.
If you think back to being 13 or whatever and just driving a cop car into into a wall. and now I can go and instead of driving a cup car into a wall, I can deadlift and adopt a dog and do all kinds of pro social things. Yeah, so just to kind of gloss that a little bit, like the first two games are top-down, then the games go 3D. And this is a period of time in which a lot of games are going from sort of these little sprites that you see in two dimensions. So think of like Zelda, the old Zelda games which are top-down and you're running a little fairy man around or like Mario, which you're seeing like, you know, like a poster on the wall, but that moves. And then they're entering three dimensions. And so like taking a game like Grand Theft Auto, which was two dimensional and very cartoonish and rudimentary, like there's sprites here that are like shooting and driving. And then doing that in a three dimensional space that is supposed to recreate New York,
you know like what's right behind me that was radical like radical and also like scared people a lot and the games were very controversial because of that because there there was this idea at the time that like how are we going to distinguish uh fantasy from reality um and it's just like you go back and it's laughable now but also if you go back and look at these games and it's like jesus how did anyone think that that was a one-to-one but like you you kind of if you put yourself in that mind frame, it's a lot easier to understand. And one of the reasons that Rockstar is so, holds information so tightly is exactly because it was so controversial back then. The context of all this is people are obsessed with Rockstar. The more secretive they become, the more obsessed gamers get, some subset of gamers get. And there's a financial reward to like leaking pre-development or like footage of the game in development, which actually happened a couple years ago, and so Rockstar's unbelievably tightly controlled about letting media escape.
Walking into the office itself is kind of like a fortress. So to play the game involved a week negotiation between the New York Times lawyers excellent lawyers and Rockstar lawyers You had to sign things Yeah yeah I had to sign And, you know, the Times, I think, doesn't usually do that, but I think the sort of stakes of the games, this huge cultural phenomenon, were such that they had this kind of limited NDA. You didn't explain to your kid that, like, if you peek into the room while I'm playing this, Daddy could go to jail. No, I literally don't own you now. I told my six and my four-year-old that I had the coolest video game ever, but they couldn't watch because it was against the law. That's a core memory right there for them. But it's also cool because, like, I'll be, you know, there's ads for the game everywhere. I don't know about in the Bay. But, like, on the subway, there's just ads for them until, like, I'll be taking my kid to jujitsu or something, and he'll be like, Daddy, there's your game.
And you have to shush him real quick on the subway I have another question about this. I want to hear from Joe. Your article has this really great description of how now that the way that this thing used to satirize our culture, but now our culture is kind of not satirizable. And it's got this sort of element of tech in it that makes it not that – It doesn't seem that exciting to me to, like, be playing a video game where my video game character is just looking at Instagram and using Venmo and, like, calling rides on Uber. Like, that just, it seems like you describe it, I think, as quotidian. And so, like, maybe you could talk a little bit about the role of tech in the game. Sure. So just to back up to the beginning of your question, the games are satire. They come of age in a time in which the dominant satirical mode is like Jon Stewart Daily Show, South Park, American Idiot.
It's like making fun of Bush era Republicans. And that was like a very well-trod thing that was very sort of like kind of easy to digest if you were in that cultural stream. One of the questions in my piece is it's been well-established. We're in a very difficult time for satire. because reality, it's very hard to stand apart from what's going on and sort of make fun of it in a way that is funnier or adds a commentary to what is already happening. Yeah, the creators of the Silicon Valley HBO show talk about this, how they won't remake it now because they don't think that it's satirizable. What could be funnier than Elon jumping up and down with his shirt coming up? There is nothing. I mean, there are funnier things than that, but you can't. The height of comedy. Although I guess Counterpoint, SNL this weekend did an okay job. Yeah, yeah, yeah. I think one of the questions I really had about this game was how it was going to sort of, because they are these sort of granular recreations of American life,
how they would reckon with that dynamic. And so one of the things they do that I think is very clever is they insert it into the gameplay. And so early in the game, you are sort of a poor criminal nobody. And the joke of the game is essentially that you're too incompetent and poor of a criminal to really steal a car. And so that it's easier for you to just like rob a Burger King and take an Uber to the next place you live. So it's like fetalizing the convenience of like all of our of our of our like sort of tech saturated lives. It's saying that even the, like, criminal escapism fantasy can no longer credibly ignore the fact that everyone is, like, oppressed by technology. Incredible. That's pretty good. I love it. It's amazing that they are – it's set in Miami at this moment, too. I mean, they must have decided to do this back in – when they first started making the game more than 10 years ago. But it's hard to think of a city that's better positioned to be the site of Miami Crypto Capital,
the place where a bunch of VCs claimed they would be moving in the midst of the pandemic. Yeah, I mean, this sort of omnipresence and sinister aspect of tech is around you in the game. There's Crypto Bros. And then the sort of Amazon equivalent in the game is called Sahara. And they drive these like, they basically look like armored personnel carriers. and, like, they're super fortified. And, like, if you try and rob one, they have, like, armed guards. So, like, the fact that you, like, live in this kind of libertarian, like, techno nightmare is definitely part of the game. But, like, when in the sort of Miami rise and fall are we? Like, are we in 2021? Is there a Keith Raboi character, like, saying everyone move here? Is, you know, the mayor trying to, like, pay all the employees in Bitcoin or all the city employees in Bitcoin? or is it kind of like, you know, when, like now, when Keith or Bois moved back to New York? So one of the things I know Rockstar dealt with is like,
how do you make topical satire when the lead time is so long? So they don't have COVID for that reason, even though you would think like COVID and... No, Miami never had COVID. I went to that Bitcoin conference in 2021. There was no evidence of it. I just want to like take a moment to think a little bit about the sort of cultural moment and the cultural impact of the game. Like, part of the reason for all the secrecy is obviously Rockstar's negative, you know, experience of this in the past, but it's also, this feels like a sort of monoculture event in a way that is really rarely kind of experienced anymore. I'm trying to think of even how, like, maybe this is a question for Mike, say. Like, why would you say people care so much? Your biggest loser on the podcast. I'm going to ask the unemployed guy. Why do people care so much? Like, why do you think people care so much about GTA? I mean, look, I would argue this is like the last big video game out there. These big budget games don't really happen in the same way anymore. And that, I would argue, is a number of different forces.
Cost too much. Consoles aren't really profitable in the same way they were. But people can, like, play, you know, Flappy Bird. I mean, this is aging me. But whatever stupid iPhone game they want. First we have Furbies. Now you're doing Flappy Bird. We really need to update Mike's references. I'm just going to send you a file with things that happened between 2018 and 2026, just so that maybe you can join the conversation. Fortnite, all right? So let's do it with Fortnite. You can play Fortnite. You can get Lady Gaga with a flamethrower jumping into the digital equivalent of a Burger King online anytime you want, right? And do that for free. And that is how people game now. And that is different than paying $100 for this huge game. So I would argue it's one of the last of its time, if not the last big video game. And I think, look, this is also like, think of it in movie world. There's enduring franchises that stick around that people get excited for,
but there's not that many of them that really stick around. And I think GTA is like the last one. And then, like, look, also, you know, if you want to get cultural critic, everything is expensive. My life is a nightmare. I don't have a job. All the Marvel movies look terrible. This is, like, the light at the end of the tunnel for me right now. I mean, it's funny you say that because you are literally describing the opening premise of the game. You don't have a job. Everything is expensive. Movies suck. There is some illitude. Why not play a game that gives you all this? So I assume that the GTA 7 is going to have a GTA, an in-game GTA that allows you to play through all these things. There are in-game video games. Oh, my God. Oh, my God. So playable in-game video games. This is the real metaverse. This is what Mark Zuckerberg should have spent his way on. Mike touched on a couple of really important things. One is that from a financial perspective, I don't think Rockstar needed to make this game because they make a lot of money on Grand Theft Auto Online. And so, yeah, this game will make a ton of money, but they were printing cash from GTA Online.
So the idea of introducing something new, exactly to Mike's point about Hollywood, like, why risk it? Why make something new when we're already making money off of whatever this is? And so I think part of the kind of, like, messianic quality of, like, wow, this game is, like, the promised one is because there hasn't been anything that feels new blockbuster and high quality in a long time. So that's the hope, right? That people still want this. I think the closest thing to it this past year was probably the Odyssey. In that it was like a huge budget thing that basically delivered. It seems like the reception to it is even more positive than the Odyssey, which is pretty exceptional. And so you've talked to them, and they're making this for the love of the art, basically. Like, not because of Rack. Don't laugh. It's not because it's real. No, I'm not. Why would I laugh at GTA 6? It's art. The businessmen are not making it for the love of them. I think there are developers at that company who genuinely believe in the form and want to create something new and exciting and unprecedented.
And, like, what more could you ask for from a media experience, you know? I mean, new and exciting and unprecedented, those are, like, three adjectives that do not describe what is on television right now, what is in movie theaters for the most part. And so I think that's, like, why people are so excited about it. to be corny, to be corny about it. You guys are, like, making me want to play a video game? Look at that. Look at that. We do it. That was the point. Well, I mean, so what I'm hearing is that there's something about, in a world where not much is happening on a cultural front, that things feel worse, chintzier, they cost more money, but there's not less money being put into making them. This is a really monumental achievement that is kind of hitting, you know, as they say in Hollywood, all four quadrants, that is like really getting out there that people really care about. And that is, as much as our cultural history with GTA, as much as whatever else, this is like the sort of last gasp of this sort of effort. I mean, it reminds me of like those great Hollywood studio movies that just before the 60s downturn,
when you have like Lawrence of Arabia and Dr. Zhivago, these like last gasp of a studio system able to create something huge and great. I do want to ask, before we get into the AI of it all, which is something that we obviously need to talk about, I want to ask one more set of questions about something that is sort of unprecedented in the GTA world and feels quite different from anything that's ever in the past. One of the central aspects of the storyline in this game is a romance. Like, part of what you're doing here is engaging in a model of healthy couplehood. Obviously, the part where you're robbing cars is quite unhealthy, but it's like a press X to hold hands kind of game. Is that right, Joe? Yeah. I mean, it is. It's a frequent prompt is to hold hands. Oh, that's not a joke. I was freaking. It's really sweet. I have a few five of... Teach the youth how to be effective. Yeah, I have two thoughts about that. It used to be like, we were scared video games were making kids more violent, and now we're scared they're teaching them to love.
It's taking a whole COVID adult generation and teaching them how to love. Right. I mean, I think that that is one take. Another take, if you're of a more advanced age, is that you're holding hands with a virtual woman instead of your partner. Are you sure? Yeah. So, yeah, there is that aspect to it. Again, there's all these sort of quotidian things. It's a Bonnie and Clyde story. And you can, when you get to a certain point in the game, you can freely switch between the two characters, which is cool because GTA has historically kind of had a problem with women. And I would say the female character in this is the more richly drawn of the two. Um, she's, has a more interesting story and is, you know, it's interesting, like when you're, there's more sort of freedom of movement in this game and there's more ways to skin a cat, so to speak. Hopefully not literally in the GTA universe. Sorry. I can't skin a cat. As far as I know, you can't skin a cat. You can, you can jet ski with the cat, but you can't skin a cat. So the way other characters respond to you changes depending on who you are.
And so it just sort of, it almost doubles, in some ways it doubles the size of the world because all of the people you cross paths with, there's like 30,000 of them in the game, all unique. They react differently to you as a Latina woman than they would as a white man. So it's kind of an interesting thing. There's role-playing, you know, in the original sense of the term. I want to move, so I do, let's get to AI because this game is coming out a week after I was on Twitter and I was watching people one-shot whole video games or levels in video games using the latest Claude model. So they would post, you know, like a browser-hosted first-person shooter that frankly looked and felt pretty professional. And it's hard not to look at something like GTA. Again, 13 years, 5,000 people. We don't know exactly how much money, but let's assume it's a bazillion dollars. Now you can do a version of that for $400 or $500 in ChatGPT or in Claude.
So how are you feeling about this shift, and where are you seeing it going? So maybe the best way to think about it is... I thought you were going to cry. I thought you were going to be a little bit at the death of the video game industry. First of all, the video game industry is full of crap that is no better than something that took a couple hours in Claude. yeah it just is it's full of a lot of bad imitative boring shit i would reframe this to use the sort of hoary framework within this sort of question of taste i think that there are a lot of people at rockstar who have really good taste not just in video games but like in life and that that is distributed through the company along with the enormous quantities of time and money that they have that ensure the product for now is going to be something that is just like obviously better than something made in Claude how much longer that is the case like at what point can Claude get to like 98 of the combined taste of every rock star employee like I don't know
that's not a question I can answer beyond like a year out I have no idea in the same way people don't know what's going to happen with AI period yeah and it's also but like isn't a big piece of the appeal of why you guys want to play this because everyone else is and you want to participate in the like kind of cultural zeitgeist in the same way that we all want to see like the big blockbuster movies like if there's all these little splinterings of it and and sort of like fan fiction versions of it created by ai like that doesn't seem like it is the same and and doesn't give you the same sort of satisfaction but you know correct me if i'm wrong i mean i i don't know I'm perhaps I'm stuck in the past as clearly as many of my dated references sort of illustrate. But I do feel like people still respond. Like we've been marveling over maybe except for Aaron marveling over the details of like we want her over. Don't don't retreat. Pick up a Coke can or whatever in these games. I still think that people really recognize work and the labor behind some of these games, at least right now, you know, quality.
Right. I mean. And, like, the distinction between good quality and no quality. And I think, like, people who have lived through this transition can make that distinction. I worry about the future. But that's a broader question than video games. It's about media in general. I know. I mean, I totally agree, but it's very funny to think about, you know, these poor kids. They never played GTA 3. They're never going to make it. They're never going to have any face. This is the most elder millennial version of this podcast we ever made and saying something I mean I was going to say Joe reading your piece a word that kept coming to mind bizarrely for talking about a Rockstar game is artisanal There is something sort of artisanal about what they were doing here. There is like a level of care and craft. And you were telling me right before we started recording that this is they've been this has, in fact, been a sort of selling point for them, that the way Take Two, the Rockstar parent company, has been talking about this game very deliberately saying there was no AI used here. Yeah, Strauss Zelnick, who's the bizarrely young-looking CEO of Take-Two.
You should really look him up. He's remarkably handsome for a 70-odd-year-old man. I hope the people say that I've met him in 30 years. He's done a media tour recently, and he's touting. I mean, it's a sales pitch, basically. But it's also a statement of fact, which is that when you play it, you can feel that it's not. And as you say, it's an ironic thing to say about a video game. it doesn't feel machine-tooled, robotic. You know, I mean, it feels like something that people have put a lot of craft and care into. Okay, I've got two final questions. Joe, this sounds like a rave review. Everybody should play GTA VI, yes? Yeah, I mean, I think it's a great game. And monumental achievement. Aaron, final question for you. I feel like we're winning you over. It sounds like maybe you want to play. Will you stream it on Twitch with me? Yes, I just, Mike, can I borrow your PS4? No, not at all. First of all, it's PS5, and second of all, I will be using it. Okay, can I say that again then? No, no, no, no, no, no, no, no, no, no, we're not cutting that out.
That's paying in. I'm going to get so attacked for that. We are going to negotiate this exclusive Twitch stream during the break. After that, Joe has to go, but Mike and Erin are going to stick around. We're going to talk about NVIDIA. We're going to talk about Jensen Wong. We're going to talk about what the hell is going on with all the networks and systems that have been going down all week. This is Hard Fork. Welcome back to Hard Fork. I'm here with Times reporter Aaron Griffith and soon-to-be Atlantic writer Mike Isaac, who is joining us from his unemployed headquarters where he's been playing through Elden Ring. We're not only going to talk about video games on this episode. We will be talking about the network outages that plagued the entire world this week and, of course, NVIDIA and the ongoing saga of Jensen Wong.
I want to note here, I said we're not only talking about video games, but NVIDIA, of course, is a graphics card company, and it furnished the guts of many, many gaming PCs, Xboxes, PlayStation 3s, Nintendo Switches. Mike, as a gamer, do you always expect NVIDIA to be one of the richest and most powerful companies in the world? I feel vindicated for my lifelong obsession with this company because I used to have to save up all of my money to buy these graphics cards, which have inadvertently become the most valuable things in the entire world. But they have been around for such a long time because of how important they were to the gaming industry. I know. It's funny that many people know what a chip-making company even is, but it's because that was the best possible cards you could get for your gaming PCs. Anyway, we're not really going to talk about NVIDIA GPUs from 2007, though I'm sure Mike would be happy to do it at length. Instead, we're going to talk about the open CEO position. They have an interim CEO, and they've got presumably some internal candidates.
There is one external candidate. That would be the businessman Jensen Huang. Oh, yeah. He filled out an application on LinkedIn. Well, he put his name forward on Twitter. I'm going to read what he said. Since my departure, I've seen an outpouring of calls from NVIDIA's employees, investors, business partners, and customers for me to return as CEO. So if the board of directors decides that my return is in the best interest of our shareholders, our employees, the company and America, I would be honored to answer these calls and return as CEO. Bringing America into it. Erin, is this a real possibility? I mean, Sam Altman left and then was forced out by the board and then came back at OpenAI. I mean, I never want to rule it out because, I mean, just a week ago, I think most commentators were saying, oh, Jensen will never be fired. And then it happened very quickly that he resigned. But, you know, the comparisons to the Sam Altman blip have been kind of interesting because it is a very, very different situation. I mean, first of all, Sam was fired for what we later found out to be no reason, really.
No good reason. I think the official reason was that he was sort of a weird liar, and they didn't like that very much. Right. Not consistently candid. Bad violence. And Jensen, I think, went beyond being not consistently candid. You know, we've seen from reporting that he lied to the board. To be fair, he was also not consistently candid, just in a sort of higher stakes kind of a situation. Yeah, that's a kind of weaselly way to say it, I guess. There's been a huge difference in the situation here, just starting with the fact that OpenAI was a private company and NVIDIA is a publicly traded company. And so that just really narrows the scope of what all of these people that he's talking about, employees, investors, business partners, can actually do because there's just legal liability here. And so, you know, the board, if they bring him back, they can't just sort of like unilaterally make this decision. They open themselves up to the possibility of shareholder lawsuits. So everything has to be so by the book. And in the case of OpenAI, you know, it's just like kind of a handful of VCs that get together in a room and they're like, all right, let's like let's just do this.
Let's override this governance board. But at Firestone Haltman, it's just not as easy this way. And I think it's also worth noting that NVIDIA is unique in the universe of even publicly traded tech companies now in the fact that they have actual real governance at their company because the company is so much older and was set up prior to this era of like founder friendly shares where like, you know, Mark Zuckerberg and Evan Spiegel and all of these like kind of tech founders have positioned themselves as like an authoritarian king over the company and not as a CEO who is beholden to shareholders. shareholders and a board of directors. Most tech companies now, including Anthropic, going public are setting up essentially like no vote shares, right, for shareholders. And so NVIDIA is not set up that way because it's from a different generation. And so there's real governance that has to play out here. And so it's very different in that way, too. It feels like such an antiquated notion having like actual governing functions of a board rather
than it being like the sort of like marionette strings of the CEO on a dual class share structure thing. When I first found out about the affairs and the sort of like tricking the board stuff, the first thing I did was like, what is the stock structure of this company? Because it doesn't matter if he controls it the same way that Zuckerberg and like basically this entire generation of founders does. But like you said, like he's at the mercy of the board of directors. And that feels it just feels like old fashioned in a way that I haven't thought about for a tech reporter in a while. This makes this debate over whether Jensen should or can come back into like almost a philosophical debate over, you know, the way that tech companies should be run. Because, I mean, Mike, you and I have written about this a ton over the last decade, this kind of shift in the pendulum swinging between founder friendliness and investors giving founders everything they want and sort of treating them like princes who have absolute authority and control over the companies they create. to, you know, investor sort of having more power and being able to fire a CEO, which is right.
And Silicon Valley goes from extreme to extreme. But I think on the whole, over the last 10 to 15 years, the industry has moved toward this kind of like founder supremacy model. And NVIDIA and the way that it's set up is just like a vestige of something that is like pretty much obsolete at this point. Yeah, though we should say that like sort of as a matter of culture, NVIDIA was very much a founder-led company. And Jensen, as we know, ran it sort of by fiat. He had 60 direct reports or whatever. He was the face of the company. He was making these sort of visionary long-term bets. And I think that sort of fed into this sense last week they'll never get rid of him. But as you guys keep saying, like clearly there is in fact like a legal structure here that would allow him to get pushed out. But I want to stay on Aaron's point for a second because I think it's worth looking at the people who have been publicly calling for Jensen's return. There was a petition going around. Elon Musk signed it. Mike's old friend, Travis Kalanick, signed it. Palmer Luckey has been calling for his return. The president appears to be on board again.
No, who can say his original suggestion was for Tim Cook to come back. Presumably he had to change his name from Tim Apple to Tim NVIDIA. So Tim Vidia. Oh, like it. Right. But like we're talking about a lot of founders and venture capitalists. And also we're sort of talking about the tech right. Right. Well, I mean, it's more than that. It's also the employees. It's the customers. It's, you know, Jensen, it's America. But it's worth, you know, digging into or thinking a little bit about why they all want him back so badly. I mean, the tech industry, regardless of their political views, relies on this one company and this one guy more than anything or anyone else. I mean, I've talked to people who called him the industry's rich uncle. He's, you know, NVIDIA is the Federal Reserve and he's the Fed chair. He's the benefactor. Like, it's not just that NVIDIA is supplying the chips. They're also one of the most aggressive acquirers of startups at very high valuations, one of the most aggressive investors in startups.
And most critically, NVIDIA has been backstopping all of these neocloud and infrastructure companies with sort of revenue guarantees. And then on top of that all, he's also the biggest booster of open source. And that is a huge piece of, I guess, that's a huge sort of like philosophical view on the way that the tech industry, the direction it's going to move in. So if there's someone who's a little bit more conservative or not as aggressive in the dealmaking in charge of the company, that's just going to create this huge chilling effect for the entire industry. And so there's a piece of that. It's not just a political thing. It's like people in the tech industry really view him as critical to the success of the future of like where everything has been going. Imagine like I'm trying to think of it. You're inspiring me to think of it as like regime change in essentially a government. Right. Like you have all of these players who have invested this much time and energy into getting getting in front of the guy who makes all the decisions, whether he's buying your startup or funding your important company.
That's going to be the next thing. And then you have someone else who comes in and is like, actually, I don't care about you or whatever. You know, it only took Mark Zuckerberg putting on a leather jacket to jump to the front of the line for NVIDIA chips hanging out with Jensen. They did a whole fashion swap. I think, like, Mark wore the leather jacket and Jensen wore, like, Mark's shearling hypebeast coat or something like that. Wow. That was the thing. That's nice. It was pretty nice. Yeah, I miss that. Pretty adorable. Yeah. I don't see Colette Kress doing that. the CFO of NVIDIA that they brought in. I mean, it's also just worth pointing out that she's already, I mean, there's been some reporting that has come out that has shown that she has pushed back against some of Jensen's really aggressive deal-making, like, over the last year, you know, starting to, like, kind of get a little bit nervous about the company getting out over at Skis on some of this stuff. And she's kind of, you know, made some comments to the effect that, like, okay, you know, we're going to maybe be a little bit more conservative going forward. And, you know, obviously we have yet to see that. But yeah.
One thing I will say, though, on your point, Max, is like there's another framework you could look at it as like this is this sort of shift from, let's say, a post or a pre midterms world where the folks in tech land who were, you know, comfortable with Trump land, who were sort of cozying up to right wing or conservative politics, who were like ascendant essentially are seeing this figurehead of that movement essentially get toppled, right? And like maybe the people who were comfortable with that or at least like on board with that are kind of like, okay, this is – it's not going in our direction anymore. So, you know, what can we do to get back towards that? And I don't know. It feels strange. It also feels like sort of like a last grasp that may not work, but they seem to be playing out on Twitter. But that's a good place to go, Mike, because we should also talk about the people who are not in favor of Jensen coming back or the people, let's say, who are sort of conspicuously silent about it. Obviously, the board has quite openly stated, Harvey Jones, the new board chair, says,
the NVIDIA board unanimously accepted his resignation, does not intend to reinstate Jensen as CEO. There's been some reporting that's kind of unclear about factions on the board who are maybe pro or anti-Jensen. I don't know if you guys have seen this, but I'm not quite sure how seriously to take that. It's just hard to know. There's so much still that has yet to come out about what he's done and what he knew. And I think it's going to be really hard to make a call on the likeliness until we actually know everything, which is actually what some of the people who are calling for his return are also saying. Like, we don't know everything yet. Yeah. But we want him in the driver's seat regardless. That's why we investigate, I guess. This is also obviously a situation where a lot of the reporting, including reporting about factions on the board and who's in support and who's against them, is part of a general campaign on either side to push for his return or to push for his permanent ousting. So there's still a sort of fog of war, to misuse a metaphor, a sort of fog of war thing right here. One thing that's very interesting and funny to me is there's a bunch of CEOs who are sort of nailing this.
We love Jensen. We support him. And I'm not going to say anything more than that. today at the New York Times Deal Book Summit Sam Altman and Greg Brockman were both asked about Jensen. Whitney has a clip of this if you can show it Whitney Do you think she can come back? Greg and I have like a little more empathy towards the situation than the average person and I will say one thing I did not appreciate was everybody on earth speculating with some random theory here or there so I will say I wish Jensen the best, I wish NVIDIA the best I don't think the world needs another set of random opinions on what's going to happen here. Oh, I feel personally attacked by that. Beautiful. Because we are offering another set of random opinions. No, no. We're the last set of random opinions. Sam would be one too many. We're the final ones. We're putting a lid on it after this. I do like, this seems to be the tactic that a lot of people I saw, Jeff Bezos was also saying something very similar this week. This sort of like, love Jensen. He's a genius. He's a visionary.
And that's all I can really say on the subject. And now let's move on to the next question. It's kind of like they almost learned from like getting too in line with the Republicans. They realized that we need to play both sides of every situation here. Like it's just good business. If it's not Jensen that comes back, whoever does come in, we want to be in with them as well. Yeah, exactly. And, you know, NVIDIA does seem, per Sam's wishes, does seem to be doing quite well. Its quarterly sales topped $100 billion for the first time. It reported in its earnings on Tuesday. So the business, you know, is pretty healthy. Yeah, and they had this all-hands on Wednesday where all the leaders, like so many people from the leadership came in, including some of the original co-founders of the company, just reassuring the troops, like, we're going to persevere, we're going to make it through this. We have an incredible business. Obviously, people still want chips. That doesn change It really the long I think vision that Jensen supplied that they going to need to replace But in the short term for the next few quarters I think they be doing great
We'll start to see the effects in many months down the road. I do think that's one of these risks around this cult of the founder, great man of history issue you have with really founder-led companies. It goes back to the debate that we were talking about earlier. Yeah, that's exactly it. And, like, that's why you have the execs at the top doing this all hands. I mean, we've seen this with so many of these weird, like, internal wars. You come in, you've got to calm the troops. There's all these people saying, I'm going to quit. You have actual talent that you don't want to leave your company. That's pretty important, I would argue. Yeah. So while a bunch of investors and executives are arguing over who gets to run NVIDIA, the rest of us have been stuck in airports, locked out of our email, unable to use our banks. I'm just going to run down a list of the companies and institutions that have faced major network issues over the past 10 days. On Thursday, Microsoft Outlook went down. On Friday, the FAA grounded every plane in the U.S. over a communications outage. On Monday, trace transactions were disrupted.
That same day, United flights were stopped after a widespread system outage. On Tuesday, Apple Pay and iCloud went down. Later that day, a classified NSA network went down, which I don't believe affected me, but who could possibly say? Did you guys get grounded? Were you stuck at airports or anything, or were you mostly unaffected? I was not. Maybe I just lived too much of an analog life, or I'm just like a hermit, but no, I personally wasn't. I've been a cocoon of video games and unemployment, so I have no idea what's going on. What I have been doing, though, is I've been reading all these prepper guides that have been circulating, like how to prepare for, you know, if the grid goes down and what you should do. And I feel like half of them are AI-generated, but are you guys getting these in your feeds? because I'm like reading it like, oh, we got to get more water. And like, I got to, you know, not my feet. Well, I have been thinking about like one of the things is like you should get a landline. And I'm like, yeah, it would be great to have a landline for when all the all the, you know, cell phone towers are down. But like, who would I call?
Let's the three of us get landlines and talk more. I think that would be really nice. Whether or not everything goes down, I think it would just be it would be nice of us to have a couple of weekly conversations on a landline. As someone who has been fully out of the loop, can you actually tell me why this is happening, Max? Like, do we know anything at this point? Yeah, there's a lot of conspiracy theories out there. Like, let's get the real stuff. Well, so the official explanation, not to sound like a conspiracy theorist, but it's clearly like a developing situation. But Senator Wark Warner, the vice chair of the Senate Intelligence Committee, has been saying on Twitter and in a front-facing video, which all senators, even the septo-tenarians, have to do, that what's been going on is all of these systems are in the process of updating because of AI discovered vulnerabilities. So over the last year, especially with the help of some of the frontier labs, a bunch of the biggest institutions in the country, places like Chase Bank, have been using AI systems to probe their networks and figure out what's going, you know,
if there are particular vulnerabilities. And obviously once those vulnerabilities are found, and it seems like many have been found, they need to update. And sometimes when you update, things go wrong, especially if you're rushing to update because you think the vulnerability is quite serious. So that seems to be what the banks and the airlines have been telling senators and the government. There was some reporting that this has been the explanation proffered. And it seems pretty reasonable to me. It's sort of interesting because I think there was a lot of speculation early on that this was like a cyber attack or this was rogue AI or whatever. It is, in fact, AI, but it's kind of an inverse. It's like AI has discovered a bunch of problems that need to be fixed, and the humans in rushing to fix those problems have tripped over themselves a little bit. I want to be clear, too, this isn't true across the board. The reporting seems to be, for example, that the Microsoft Outlook bug was both entirely human-created and mostly human-fixed, that it was created by humans, it was discovered by humans, and then they used AI to help fix it. So that wasn't one of these big system updates that fell on its face. But this is the general thing. I just feel like everyone is like really on edge and like prime to assume like everything is this nefarious thing that's coming after us where I don't know like having reported on some of this stuff for a while maybe I'm being naive but I do feel like the kind of Hammond's razor dumbest possible like version of what happened can be accurate a lot of the time like sure we all turned it off and on at the same time and it happened to like break everything kind of makes sense to me.
I don't know. You seem a little skeptical, Max. I mean, it's just another sign, I think, of the lack of trust in the tech companies. I mean, it's similar to like, you know, when something happens in politics, people create these conspiracy theories because they don't trust politicians. Like, I think, you know, the tech industry is finding itself in the same position where whatever the explanation might be, people are not going to believe it because they don't trust these tech leaders. I mean, it's not just not trusting. It reminds me a little of the data centers and politics stuff where it's like, you know, if you've been paying even a little bit of attention for the last couple of years, you've heard people at the AI labs talking constantly about the dangers to infrastructure emerging out of the models they're creating. So why wouldn't your mind immediately go to that? You know, we are we are you know, it makes perfect sense to me that that's the first thing people would think. And so I get the feeling this is not going to be the last time we see a spurt of outages like this that have their own individual causes that can be attributed in a pretty straightforward and transparent way. But the people's first thought is going to be, oh, well, you know, it's Iranian or Chinese AI or it's a rogue AI or it's, you know, it's script kitties, you know, who have figured out how to get it, all this stuff.
It feels like this is kind of a picture of the next few years of our lives, hopefully only the next few years of our lives and not the rest of them. But it's hard to think it's going to go away. I mean, one. That was sort of my question is like, is it just that, OK, AI is going to discover all the vulnerabilities and so we have to patch it all. And then like we're going to go through this like kind of ugly period and then we'll go back to just normalcy. Or are we just going to live our lives not feeling like any of our basic infrastructure that we've all ported online will ever be stable again? You guys should stay tuned to Hard Fork because I know in a future episode we are definitely going to be talking about this. There is some really interesting and, frankly, quite terrifying reporting being done on the quality of, like, infrastructural hardware, software, and firmware, you know, like water desalination plants, power plants, all this stuff that is really not in a great position to be defended. We'll be delivering that episode of Hard Fork over landlines. Exactly. So how are you guys feeling a little nervous about flying for the holidays? Oh, no. Oh, no. Not until right now.
Thank you. Okay, I'm going to leave us. I guess I should be. I'm going to leave us on that. Hopefully, we all make it. Hopefully, everybody is listening to this not over ham radio or something. This has been Heart's Work. Aaron and Mike, thank you so much for joining us. Thank you. Yeah, thanks for having me. We're going to come back after the break. Aaron's going to join us again, and we're going to be talking about the Amthropic IPO and whether or not $2 trillion is a lot of money. We're back with Aaron Griffith. We're recording this on Thursday, the day of Anthropics IPO, just moments after stock began trading at $111.50 per share for a fully diluted valuation of more than $2 trillion. Erin, my first question, is that a lot of money? Well, you know, in the grand context of how much has been dumped into the AI boom over the last four years, maybe not.
But for a five-year-old startup going public, yes, that is an absurd, obscene amount of money. And it is going to be, you know, raining cash all around San Francisco and Silicon Valley and maybe, you know, a data center near you. What are your big takeaways from this IPO, from this figure, and what you're seeing out there? For one thing, they're defying all of the macroeconomic and geopolitical and kind of market jittery fears that have been swirling in the market and kind of like building on one another throughout the year. I mean, we had the fears of the AI bubble and the infrastructure buildup. We had the midterms where, you know, the country basically said we don't like AI or any of the people building it. And we especially hate data centers and the infrastructure that you need. And then we have this NVIDIA scandal and Jensen Wong getting ousted and, you know, a lot of questions around the, quote, unquote, Fed chair of the of the AI boom. And so amid all of that, it seems a little bit insane that a five year old startup that, you know, came out of nowhere, basically, is now worth roughly two trillion dollars and was able to raise more than 200 billion dollars from public market investors.
and is now trading on the stock market. So that piece of it is kind of what I'm marveling at this morning. Right. So it seems fair to say if there is an AI bubble, this IPO is not the thing that's going to pop it. That is the open question. I'm kind of wondering if a lot of these public market investors who are just scrambling to get into this IPO, they are so excited to have the chance to invest in a pure play IPO, like an LLM company that's the frontier lab at the cutting edge of this tech. if they're selling off all their other AI investments in order to pile into Anthropic. That's one thing I've kind of heard and wondered about. Unclear if that's the case or not. Or if Anthropic is going to be kind of the rising tide that lifts all the other boats. I don't really know. Right now, investors are, you know, just kind of taking a victory lap and talking about how smart they were to get in on this company, even though many of them, most of them, passed on early opportunities to invest in Anthropic and then only piled in later when the company was obviously a, quote-unquote, consensus bet,
like it was an obvious winner. And now they're all, like, very excited and patting themselves on the back. And I think that's going to continue. And a lot of them have, you know, invested even more in the IPO. So just from an investment perspective, it does seem like investors think the AI trade has a lot more room to go. And what kinds of people and institutions are buying the stock today? Is the demand institutional? Is it retail? What are we seeing? I mean, this is a very interesting piece of these mega IPOs that we're starting to see now. I mean, we saw this with SpaceX, and it's been kind of a continuing trend for a little while. But, you know, when a really big IPO that has a lot of name recognition, a lot of brand behind it, obviously Elon Musk is the prime example of this, but Anthropic and probably next year OpenAI as well, will be, they are now trying to find ways to get retail investors bought in because they're the ones that are kind of promoting and talking about the stock on Reddit and kind of making it into like maybe a meme stock, maybe kind of borderline. But anyway, retail is an increasingly important component of any IPO and any kind of public
market investment composition strategy. And so Anthropic did that as well. They've talked about how it's kind of part of their mission. and they want to bring everyone along because this is the age of miracles, and maybe we might all be out of jobs because of AI soon. So why not buy our stock? You mentioned OpenAI. What do we think this tells us about their expectations and schedule for them to go public? OpenAI is supposed to go public next year. Obviously, this shows there's a ton of demand for a pure play LLM Frontier Lab. They're Antipak's biggest competitor. I don't think the businesses are apples to apples. OpenAI has spent a lot of time and probably gone down some kind of dead ends with consumer-facing products, and Anthropic has focused on enterprise from the very beginning. And so they have a very strong business, and they're also maybe a little bit more efficient with their use of compute and their use of capital. So investors probably have a stronger case for why they're more excited about Anthropic. But OpenAI could also play into that retail and consumer piece of it as well.
So, yeah, they're planning to go next year as far as I'm aware. We'll see how the market shakes out. I should mention by way of disclosure that the New York Times is suing OpenAI and Microsoft over alleged copyright violations related to the training of large language models. Okay, Erin, I think I have just one more question, which is looking at this IPO, $2 trillion, it's huge. Does this feel like the machine god is coming, or does this feel like a normal IPO? This has actually been kind of a disorienting experience this morning. I've been talking to all these investors. I've been, you know, watching Dario's speech. She gave this speech before ringing the opening bell on NASDAQ this morning. And all of the conversation around it feels like, oh, this is a normal IPO. This is a normal tech company. I mean, at one point, Daniela Amadei, the president, Dario's sister, did say something like, we started this company because we care about safety. That's our main concern. And we believe this technology is different than other technology.
But all the other conversations I'm seeing, you know, it's not about the machine god. It's all about ARR and credit ratings and cash reserves and all of the normal stuff that you'd think about when you're talking about an enterprise tech company. So it has been a little bit disorienting that way because we also know that this morning, Jacob Coxon, the former Anthropic employee, is staging a protest in front of the Anthropic office in San Francisco, calling on the company to ban RSI. And there's this whole conversation basically started by Anthropic swirling around about how recursive self-improvement is potentially going to be the end of us all and we're going to hand over our control of humanity. So it is extremely disorienting. I am very much wondering how these earnings calls are going to go. Every time the founders have been asked about this kind of doomer and safety question, they have a very pat answer that I feel like has probably been coached by the bankers to say,
like, yes, we care about safety, but back to our revenue and our growth. All right, Erin, thank you so much for joining us to talk about a $2 trillion company that may or may not end the world. This has been Hard Fork for this week. Thank you for listening, and we'll see you next week. This episode of Hard Fork was produced by Whitney Jones and Sawyer Roquet and edited by Viren Pavich. With additional editorial support from Brendan Klinkenberg, Lisa Tovin and Brooke Minters. We were fact-checked by Will Peischel and engineered this week by Katie McMurrin. Original music by Alicia Vietoop, Marion Lozano, Rowan Nemistow and Dan Powell. Additional video production by Pat Gunther. You can watch this full episode on YouTube at youtube.com slash hardfork. Special thanks to
Willie Staley, Sam Dolnik, Jessica Testa, Blue Wing Tam, Matty Maciello, Zach Caldwell, Luke Mylander, Chris Moore, Michael Cordero, Thomas Trudeau and Dalia Fadad. we love to hear from our audience really if you have any questions or comments or even if you want to give us a piece of your mind you can find us at newyorktimes.com and you can email us at
番組の概要欄(原文)
Anthropic goes public, Jensen Huang wants back in at Nvidia, and software patches seem to be breaking everything. But first, let’s talk about G.T.A.
出典・この記事について
- 番組
- Hard Fork(The New York Times)
- 元の回
- G.T.A. VI Is a Monumental Human Achievement
- 配信日
- 要約作成日
- 公式サイト
- www.nytimes.com/column/hard-fork
この要約・解説記事は、元の音声を文字起こしした内容をもとに AI(Claude)が日本語で作成したものです。 固有名詞や数値に誤りを含む場合があるため、正確な内容は元の音声でご確認ください。制作方法について
関連エピソード

Anthropic の1.4兆ドルIPO:人類史上最大の富の創出イベントか
Moonshots with Peter Diamandis

AI業界の行方を左右する5つの問い:Anthropic IPO、OpenAI新モデル、ジェンスン・フアンの去就
The AI Daily Brief: Artificial Intelligence News and Analysis

ジェンスン退任とNVIDIAの後継問題、AI終末論への反応、ハサビスのRSI警告|MOONSHOTS #311
Moonshots with Peter Diamandis

特別レポート:ジェンスン・フアンの失墜とその後に来るもの
Big Technology Podcast
Jensen Huang の NVIDIA CEO 辞任と、Claude Fable 5.2 は AGI なのか
Mixture of Experts

ジェンスン退任? Fable 5.2とAnthropicのIPO、Gemini 4 CarbonとHydrogen――AI Engineerからライブ
ThursdAI - Highest signal weekly AI news show