The AI Daily Brief: Artificial Intelligence News and Analysis - Sam Altman Wants to Raise $5-$7 TRILLION for AI Chip Plants

Episode Date: February 10, 2024

WSJ writes about an incredibly ambitious plan to transform the landscape for compute. Additionally, the White House has started an AI Safety Consortium. Today's Sponsors: Notion - Notion AI. Knowledge, answers, ideas. One click away. - https://notion.com/aibreakdown ABOUT THE AI BREAKDOWN The AI Breakdown helps you understand the most important news and discussions in AI.  Subscribe to The AI Breakdown newsletter: https://theaibreakdown.beehiiv.com/subscribe Subscribe to The AI Breakdown on YouTube: https://www.youtube.com/@TheAIBreakdown Join the community: bit.ly/aibreakdown Learn more: http://breakdown.network/

Today on the AI breakdown, Sam Altman is looking to raise the equivalent of 8% of global GDP, 5 to 7 trillion, yes trillions with a T, to build a global network of semiconductor plants that would radically increase our access to compute. And before that on the brief, the White House announces a new AI safety consortium. The AI breakdown is a daily podcast and video about the most important news and discussions in AI. Go to Breakdown.network for more information about our YouTube, our Discord, and our newsletter. Welcome back to the AI breakdown brief, all the AI headline news you need in around five minutes. Well, once again, we are seeing how increasingly significant AI is in the political landscape,

with the latest news being that the White House has announced a new AI safety coalition that they're calling the AI Safety Institute Consortium. It is apparently a coalition for the safe development and deployment of generative AI, and includes not only the usual suspects from the AI space, the Microsoft's, Google, Apple's, as in Open AIs, but also a number of government agencies, a set of academic institutions, and even other companies outside of AI like Northrop Grumman, BP, Qualcomm, and MasterCard. The consortium was announced by Secretary of Commerce Gina Raimondo, who said in a statement, the U.S. government has a significant role to play in setting the standards and developing the tools we need to mitigate the risks and harness the immense potential of artificial intelligence.

President Biden directed us to pull every lever to accomplish two key goals, set safety standards, and protect our innovation ecosystem. That's precisely what the USAI Safety Institute Consortium is set up to help us do. Overall, they say that the consortium includes more than 200 member companies and organizations. And it really is basically everyone. Accenture, Adobe, AMD, Bank of America, Booz Allen Hamilton, BP, Canva, Carnegie Mellon, City Group, etc., etc., etc. I won't go through the entire alphabet.

Now, it is not exactly clear what this consortium will actually do. They say that they're tasked with working on the priority actions that were outlined in the executive order from last year. That includes, quote, developing guidelines for red teaming, capability evaluations, risk management, safety, and security, and watermarking synthetic content. But what each of these members will contribute to this consortium, how the consortium will be run, who will be leading it, who will be prioritizing these things, et cetera, et cetera, et cetera, is all a little bit open right now. Then again, that's probably because they're not sure yet. Now, meanwhile, over in the UK, the AI Safety Institute there has just published its third progress report.

Chair Ian Hogarth wrote about a few different elements. First, they are expanding the team. They've announced a few key executive hires, as well, they say, as onboarding 23 technical researchers, with an aim to grow a team of 50 to 60 by the end of 2024. They published the principles behind the International Scientific Report on Advanced AI Safety, which they said was a landmark paper sharing the latest research and opinions

from the world's leading AI experts, including Turing Award winner, Joshua Benjillo, and they say they've begun pre-deployment testing for potentially harmful capabilities on advanced AI systems. Now, they say that that testing is going to focus on a few different areas. The first is misuse, or as they put it, assessing the extent to which advanced AI systems meaningfully lower barriers for human attackers seeking to cause real-world harm. They have specific interest in chemical and biological capabilities as well as cyber offense capabilities. Next, they want to focus on societal impacts, including, quote, the extent to which people are

affected by interacting with such systems. A third focus is autonomous systems, specifically the ability for systems to autonomously replicate, deceive humans, and create more powerful AI models. And finally, a fourth focus is safeguards. They also flagged that there would be another set of upcoming AI safety summits, with one happening in South Korea and a third expected to happen in France. Now, the Department of Commerce wasn't the only group talking about AI in the U.S. government. The FCC has officially made AI-generated voices in robocalls illegal. This is, of course, a response in part to that AI-generated president.

and Biden that was calling New Hampshire Democrats a couple weeks ago, but that was really just a catalyst to get moving on an issue which everyone already knew was going to be an issue. Said FCC chairwoman Jessica Rosenworsal in a statement, bad actors are using AI-generated voices in unsolicited robocalls to extort vulnerable family members, imitate celebrities, and misinformed voters. We're putting the fraudsters behind these robocalls on notice. Now, this ruling was adopted by the FCC unanimously, and in addition to making this type of activity illegal, also apparently gives state attorneys general what they call new tools to crack down. Finally, today, we have to talk about ARM. This is, of course, a British chip designer that is

majority owned by SoftBank, and in their recent earnings report, they announced that they made a record $824 million in sales last quarter and projected as much as $900 million for this quarter. Both figures were significantly ahead of what Wall Street analysts expected, and the stock price has gone up 48% since this earnings report was released. The company's executives explicitly called out AI as having given the results a boost. Now, one question is whether this is an area where the market is getting out ahead of itself. The Wall Street Journal published a chart that showed a number of different companies in terms of their share price as a multiple of forward earnings. Nvidia sits somewhere around 35, AMD sits somewhere around 45, and ARM is now up approaching 100.

The WSJ asks, Army's doing very well these days. but is it doing three times as well as NVIDIA? And this type of thing is the reason why, despite the fact that AI actually is showing up in company performance already, there's still some people who think that the market is getting ahead of itself. I think in some ways we are a little bit trapped by priors and unable in general to imagine that both might be true simultaneously, that AI is in fact every bit as disruptive as it seems, and that indeed it's more economically relevant earlier than other previous technology cycles, but there could still be on top of of that, a little hint, a whiff, if you will, of irrational exuberance. Of course, to me, it doesn't

seem like it's going to slow down anytime soon, but we will have to see. However, that is going to do it for today's AI breakdown brief. Next up, the main AI breakdown. A quick message before we get back to the episode today. At this point, you guys know that Notion is one of the major tools that I use day in and day out across the breakdown network, the AI education beta project, basically anything that I'm doing in any sort of professional or entrepreneurial endeavor is going to be anchored by Notion. Now, you also know that one of the big themes that I keep talking about for 2024 when it comes to artificial intelligence is the integration of AI into our workflows. I think in many ways it's not just about which third-party AI tool is best for any given use case, but how they actually fit into what we're doing in ways that are actually time-saving. And that's why I love that Notion now has AI so deeply integrated across its entire,

suite of tools, which means that it's everywhere in your entire workspace. Now, for those of you who don't know, Notion combines your notes, documents, and projects into one space that is simple and beautifully designed. It's your one place to connect teams, tools, and knowledge, so you can do your most meaningful work. Unlike other solutions, it doesn't have you bouncing between six different apps. It is seamlessly integrated, infinitely flexible, and incredibly easy to use. Now, with the new, fully integrated Notion AI, you can work faster, write better, think bigger, and take care of tons of tasks that might normally take you minutes or hours in just seconds. One of my favorite use cases is to use Notion for brainstorming. So, for example, what would a great launch strategy be for some

project? Use Notion AI to help you think through all of the different dimensions of how you could tell that story. Now, the proof is in the pudding and Notion is used by over half of Fortune 500 companies. And most importantly, probably for you guys, the teams that do use Notion, send less email, cancel more meetings, save time searching for work, and reduce spending on tools. Right now, you can try Notion for free when you go to notion.com slash AI breakdown. That's all lowercase letters, notion.com slash AI breakdown to try the powerful, easy to use notion AI today. And of course, when you use our link, you're supporting the show. One more time, that's notion.com slash AI breakdown. Welcome back to the AI breakdown. Siki Chen tweeted this morning, you think you're pretty ambitious,

and then you see Sam Altman fundraising for like 10% of global GDP. The story that he's referencing came out in the Wall Street Journal late last night and is called Sam Altman seeks trillions of dollars to reshape business of chips and AI. So let's go back and give this a little bit of context and then just talk about what the new information suggests about Altman's ambitions as well as the larger landscape for AI and society. When Sam Altman was being fired from OpenAI, we never really got a good justification or explanation of what the reason was. The Open AI board, of course, said that they had lost confidence in him, that they didn't think he was being truthful, and it appeared later on that perhaps part of the issue was around how Sam was using

the OpenAI name to potentially raise money for other non-open AI projects. Now, this would later be something that Sam Altman would strenuously deny, saying very clearly that everything was related to his core project, which was OpenAI. But one of the things in question was a big AI chip initiative. Now, it is of course no secret that the availability of chips has become one of the major constraints in the development of artificial intelligence. This is why we're talking about ARM stock going up 50% in a day. It's why Nvidia has been one of the best performing stocks of the last year. And over the course of the end of last year, we got a number of different reports about

conversations that Altman was having, particularly with Middle Eastern funders, as well as soft banks Masayoshi Sun, about the potential for funding some new big chip initiative. That was about all the information we had. There were some intimations that Altman wanted to focus on the U.S., although he was thinking in terms of a global network of chip fabrication plants. And there were also some indications that he was talking with TSM about potentially running the firms. Now, what wasn't clear was how much this was meant to be some big competitive endeavor to undercut or take out NVIDIA. It was sort of presented like that in media, which makes sense given that every big tech company is also racing to build their own chips,

so they're not totally dependent on NVIDIA or AMD. And so I think that some people assumed that Altman was just coming for NVIDIA and Jets and Hwang. The new reporting from the Wall Street journal paints a very different picture that, while again, we don't have anything confirmed from Altman himself, strikes me as much more resonant with the approach that it seems likely that he would take. The WSJ piece kicks off. Sam Altman was already trying to lead the development of human level artificial intelligence. Now, he has another great ambition, raising trillions of dollars to reshape the global semiconductor industry. So the big new thing here, like everyone noticed right up front, is the absolutely eye-popping amount of money that Altman is apparently talking about.

$5 to $7 trillion. Global GDP is around $84 million, so we're actually talking about 6 to 8% of global GDP for this project. Now, to put this in context of the current semiconductor industry, last year, global chip sales were $527 billion. That number is expected to rise to $1 trillion annually by 2030, and on top of that, semiconductor manufacturing equipment was another $100 billion industry last year. That means that Altman is talking about 10x roughly, the current size of the entire semiconductor industry for this project. Once again, as the WSJ writes, the amounts Altman has discussed would also be outlandishly large by the standards of corporate fundraising, larger than the national debt of some global economies and bigger than

giant sovereign wealth funds. Once again, the reference points they draw on are corporate debt issuance in the U.S., which totaled $1.44 trillion last year, and the combined market cap of Microsoft and Apple, who are the two biggest businesses in the world, which equals $6 trillion. Now, if you watch Altman at all closely, it's very clear that he has already. zoomed out to a world in which the two biggest bottlenecks on the abundance that he sees coming from AGI are compute and energy. In fact, he said exactly that much in conversations at Davos earlier this year. Now, on the energy side, in a number of startups that are focused on things like nuclear fusion, he said explicitly again at Davos that he believes that an energy

breakthrough is going to be needed to achieve the energy profile that we need for advanced AI applications. It was really interesting, actually, when he was asked about whether he was concerned about the environmental impact of all of this new AI compute power being brought online, he said he wasn't because it, by definition, required an energy breakthrough, which would make that not an issue. It's clear that this effort is focused on the other side of that equation, access to compute. And what I believe is made clear from the scale that he is apparently talking in, is that this is absolutely not in any stretch of the imagination, an endeavor that Altman is imagining as somehow him or OpenAI competing with Nvidia and trying to beat them out in the market as it's

currently organized. Instead, it reads much more to me like a global Manhattan project or a global martial plan, I don't know, pick your analogy from around the World War II period, to get everyone aligned, and I mean everyone, funders, industry, governments, to build the capacity to have as much compute as the world is going to need and can make good use of. Now, certainly it appears that he's thinking in those governmental terms. According to people familiar with the matter, he recently met with Commerce Secretary Gina Raimondo to discuss this. In other words, this isn't Sam running off to the Middle East. For some Abu Dhabi fantasy, he's thinking about the depth of partnerships that it would take at the highest governmental levels. OpenAI didn't deny that these conversations are happening.

A. Spokeswoman said, Open AI has had productive discussions about increasing global infrastructure and supply chains for chips, energy, and data centers, which are crucial for AI and other industries that rely on them. We will continue to keep the U.S. government informed, given the importance to national priorities, and look forward to sharing more details at a later date. So again, to me, this very clearly is not another new business line for OpenAI. This is OpenAI trying to use the leverage that it increasingly has to be a catalytic force for fundamentally changing the compute landscape. And the piece also did come with more confirmation about who Allman has been talking with. The Commerce Secretary part was a new detail, but it does appear that there have been lots of funder

conversations in the UAE, that soft bank has been involved in the discussions, and that TSM is being looked at as a potential partner to actually build dozens of chip fabrication plants across the world. Now, one blocker is that while Altman would like to have many of these plants be built in the U.S., where, of course, the U.S. government is already investing in expecting to invest even more into chip manufacturers to fund new factories, there are some fairly significant challenges. As the WS.J writes, those companies face challenges expanding in the U.S. TSM, for example, is pointed to delays, a shortage of skilled workers, and high cost at its $40 billion project in Arizona. There are also the political issues of getting this deeply involved with funders from the Middle East.

We've profiled in the show about how controversial G42 has been. G42 is an Abu Dhabi-based company that partnered with OpenAI back in October, but which has also been the subject of some serious scrutiny for members of Congress around its potential ties to the Chinese Communist Party. The point being that this is an incredibly ambitious plan that is challenging from the standpoint of not only capital, not only talent, but politics as well. Indeed, it may be even more ambitious in some ways to get the amount of governmental cooperation that would be required for this, as it would be to go out and find 8% of global GDP to fund it. Still, if you had any doubts about the scale of ambition of Altman and the folks who are most at the cutting edge of the artificial intelligence movement. The story, I think, shows just how world-changingly, let's say, they are

thinking. I will, of course, bring you any new details as they come up, but for now, that will do it for the AI breakdown. Until next time, peace.