Stargate: A $500B US AI Initiative Announced by Altman, Ellison and Trump
16:06 Share

Thank you.

Welcome back to the AI Daily Brief Headlines Edition, all the daily AI news you need in around five minutes. One of the big conversations heading into this year is whether we're going to see an uptick in IPOs. The last couple of years have been incredibly bad for initial public offerings. A word like anemic doesn't even really do it justice, but there is a sense that perhaps this year we will see some shift in that. And of course, one of the big drivers is enthusiasm for artificial intelligence companies.

When it comes to which companies are likely to go public, Morningstar just today pointed to six potential IPOs to watch, including one company, Klarna, that while not an AI company itself, it is instead a buy now, pay later company, still has made headlines and has made it very clear that AI is a chief part of why it's an attractive company in terms of the way that they've saved costs and updated their processes. And then two of the companies on Morningstar's list are actually just straight up AI companies in Cerebra Systems and Coreweave.

Cerebrus is a chip design company, while CoreWeave rents computing power to smaller companies.

In new news, however, it appears that French AI lab Mistral is planning to go public as well. This is a company that I think many people have seen as a likely candidate for an acquisition. When it was first launched, Mistral had a ton of energy and attention, but more recently has started to fall behind. Or at least that's the perception. Mistral has long argued that their competitive advantages include being able to run models more cheaply than rivals,

as well as work with European firms, the company being based in France, who are worried about their data being handled by firms outside the European Union. Now, whether this was just an errant comment, indeed, Arthur Mensch was specifically asked whether an IPO was in the works, and he responded, of course, that's the plan, or something more intentional, we'll have to wait and see.

Next up, Perplexity have launched Sonar, their new real-time search API. In a blog post, the company wrote, while most generative AI features today have answers informed only by training data, this limits their capabilities. To optimize for factuality and authority, APIs require a real-time connection to the internet with answers informed by trusted sources.

Perflexity claim their model outperformed leading models from Google, OpenAI, and Anthropic on benchmarks that measure factual accuracy. API access is also priced below competitors at $5 per thousand searches. The offering includes the choice of a lightweight standard model or a beefier pro model.

The API will also produce citations as an additional point of differentiation for use cases that require verifiable information. VentureBeat wrote, The move signals a significant shift in the AI landscape, as perplexity, now valued at $9 billion, directly challenges larger competitors by making its real-time web-connected search capabilities available to developers and enterprises.

Perplexity CEO Aravind Sridharvas, never short of an ambition, tweeted, The excitement around trying to make agents and vertical answer engines and legal health travel, academic patents, etc. happen is pretty clear. We want to support it through Perplexity's search-grounded LLM API, Sonar.

Lastly today, Goldman Sachs has unveiled their organization-wide AI assistant. The AI has been trained to assist bankers, traders, and asset managers. Chief Information Officer Marco Argetti said this is just the first stage in a program that will yield AI with the traits of a seasoned Goldman employee. He added, "...think about all the tasks that you might want to complete with regards to a variety of use cases for all those professions that can now be at your fingertips."

The assistant has been rolled out to 10,000 employees with the goal of making it available to all of the company's knowledge workers. For now, the assistant is capable of summarizing or proofreading emails or translating code from one language to another. Trained on internal data, Arjetty says, the AI assistant becomes really like talking to another GS employees. The next stage is agentics, with Arjetty stating, as we progress, the second step is when you're starting to have this agentic behavior. That is, I'm

Arjetty said,

For the AI to have a very specific identity that reflects the tenets, the values, the knowledge, and the way of thinking of the firm is extremely important. Now, when it comes to the idea of AI replacing or augmenting their workforce, Argetti said, "...the importance of having a phenomenal human workforce is actually going to be amplified. In my opinion, it always boils down to people. People are going to make a difference because people are going to be the ones that actually evolve the AI, educate the AI, empower the AI, and then take action."

Now, I have to say, this is one of the first times that I've seen this type of story that it actually just felt dated. I think it makes more sense as a path to an agent. And certainly there are companies that are far behind this when it comes to rolling out an AI assistant for their employees.

But my response to this really reinforced how much this is just table stakes now. I think we can officially say that if your employees do not have access to an AI assistant at this point, whether it's an off-the-shelf chat GPT or something that was custom rolled, you are as an enterprise behind.

And on that note, we will end the headlines. Next up, the main episode. Today's episode is brought to you by Vanta. Trust isn't just earned, it's demanded. Whether you're a startup founder navigating your first audit or a seasoned security professional scaling your GRC program, proving your commitment to security has never been more critical or more complex. That's where Vanta comes in.

Businesses use Vanta to establish trust by automating compliance needs across over 35 frameworks like SOC 2 and ISO 27001. Centralized security workflows complete questionnaires up to 5x faster and proactively manage vendor risk. Vanta can help you start or scale up your security program by connecting you with auditors and experts to conduct your audit and set up your security program quickly. Plus, with automation and AI throughout the platform, Vanta gives you time back so you can focus on building your company.

Join over 9,000 global companies like Atlassian, Quora, and Factory who use Vantage to manage risk and prove security in real time.

For a limited time, this audience gets $1,000 off Vanta at vanta.com slash nlw. That's v-a-n-t-a dot com slash nlw for $1,000 off. If there is one thing that's clear about AI in 2025, it's that the agents are coming. Vertical agents buy industry horizontal agent platforms.

Agents per function. If you are running a large enterprise, you will be experimenting with agents next year. And given how new this is, all of us are going to be back in pilot mode.

That's why Superintelligent is offering a new product for the beginning of this year. It's an agent readiness and opportunity audit. Over the course of a couple quick weeks, we dig in with your team to understand what type of agents make sense for you to test, what type of infrastructure support you need to be ready, and to ultimately come away with a set of actionable recommendations that get you prepared to figure out how agents can transform your business.

If you are interested in the agent readiness and opportunity audit, reach out directly to me, nlw at bsuper.ai. Put the word agent in the subject line so I know what you're talking about. And let's have you be a leader in the most dynamic part of the AI market. Hello, AI Daily Brief listeners. Taking a quick break to share some very interesting findings from KPMG's latest AI Quarterly Pulse Survey.

Did you know that 67% of business leaders expect AI to fundamentally transform their businesses within the next two years? And yet, it's not all smooth sailing. The biggest challenges that they face include things like data quality, risk management, and employee adoption. KPMG is at the forefront of helping organizations navigate these hurdles. They're not just talking about AI, they're leading the charge with practical solutions and real-world applications.

For instance, over half of the organizations surveyed are exploring AI agents to handle tasks like administrative duties and call center operations. So if you're looking to stay ahead in the AI game, keep an eye on KPMG. They're not just a part of the conversation, they're helping shape it. Learn more about how KPMG is driving AI innovation at kpmg.com slash US.

Welcome back to the AI Daily Brief. Today we are talking about a very cool one, a culmination of lots of conversations that have been happening for some time now. And the TLDR of it is that yesterday OpenAI CEO Sam Altman, Oracle CEO Larry Ellison, SoftBank's Masayoshi Sun,

came together with President Trump on his first day in office, no less, to announce a $500 billion AI data center company. So let's talk about what this project is, where it's coming from, and try to wrap our heads around the sheer magnitude of that number.

This project is all about building out U.S. data center infrastructure. The new construction will be conducted by a joint venture called the Stargate Project, which as many have noted is probably OpenAI's first actually genuinely cool name for something. In addition to OpenAI, Oracle, and SoftBank, Abu Dhabi AI fund MGX will participate as an equity partner, while ARM, Microsoft, and NVIDIA are all listed as tech partners.

Aside from the announcement being made at the White House, it doesn't appear that there's anything specific that the U.S. government has to do with this initiative. Or at least it doesn't appear that there's going to be any U.S. government money poured in. However, President Trump's interest in the project is pretty clear, with him serving not only as master of ceremonies for the announcement, but saying that the project would create over 100,000 American jobs and secure American leadership in AI. In a joint statement, Oracle, OpenAI, and SoftBank said, this

The Stargate Project is a new company which intends to invest $500 billion over the next four years building new AI infrastructure for OpenAI in the United States. We will begin deploying $100 billion immediately. This infrastructure will secure American leadership in AI, create hundreds of thousands of American jobs, and generate massive economic benefit for the entire world. The project will not only support the reindustrialization of the United States, but also provide a strategic capability to protect the national security of America and its allies.

So let's talk about some of the details that we know so far. The project will be conducted in several stages, with the first phase costing $100 billion and already being underway. This part involves the construction of a supercluster in Abilene, Texas this year. The facility seems to be the same one that was being discussed towards the end of last year as reporting emerged that OpenAI was hitting the limits of compute available for Microsoft.

The information reported that the site would cost around $3.4 billion for construction and scale to nearly a gigawatt of electricity usage by mid-2026. That would be around five times the size of XAI's Colossus supercluster, one of the largest AI supercomputers currently.

The Abilene facility is already huge, with Larry Ellison commenting, each building is a half a million square feet. There are 10 buildings currently being built. But it appears the plan is to go much, much bigger. $100 billion in data center construction would be the largest annual capital expenditure for any AI company to date, topping Microsoft's $80 billion spend for this year. The project is then planned to scale up to as many as 20 data centers throughout the country.

OpenAI researcher Noam Brown commented, This is on the scale of the Apollo project and Manhattan project when measured as a fraction of GDP. This kind of investment only happens when the science is carefully vetted and people believe it will succeed and be completely transformative. I agree it's the right time.

For reference, the Apollo program costs $280 billion in inflation-adjusted dollars, while the Manhattan Project costs $35 billion. The other big comparison is the National Highway System, which cost around $650 billion across more than a decade. Project Stargate would represent about a half a percentage point of U.S. GDP each year, similar in relative scope to these previous national megaprojects. Sam Altman took the chance to be a little bit poetic, tweeting, "...build monuments in the desert."

Now, we don't have any firm idea of how large these superclusters will be, but presumably much larger than Colossus. Colossus was already banging up against the limits of networking technology, and was frankly assumed to be impossible when construction began.

That implies that Project Stargate won't just involve unprecedented amounts of capital, but also significant technical breakthroughs. Now, in the announcement, in terms of division of responsibility, they said that SoftBank and OpenAI are the lead partners for Stargate, with SoftBank having financial responsibility and OpenAI having operational responsibility. Masayoshi Son will be the chairman of Stargate. And as part of Stargate, Oracle, NVIDIA, and OpenAI will closely collaborate to build and operate this computing system. So obviously, we'll be watching closely to see what that computing system actually consists of.

Overall, this is explicitly the project to build AGI. Altman clearly felt there needed to be a grounded explanation for the goal. He said that the technology would be able to, quote, "...cure diseases at an unprecedented rate. We will be amazed at how quickly we're curing various cancers and heart disease."

What this will do for the ability to deliver very high-quality healthcare and the cost, but really to cure the diseases at a rapid rate, I think will be among the most important things this technology does. Larry Ellison expanded on the point, stating that the technology will, quote, allow personalized medicine. Yes, it takes a huge investment, but the results of the investment will be vaccines that prevent cancers. He also discussed rapid personalized mRNA vaccine development for pandemic prevention.

Now, while this is clearly an exciting project, not everyone is necessarily enthusiastic. Noted OpenAI rival Elon Musk posted, they don't actually have the money. SoftBank has well under $10 billion secured. I have that on good authority.

This Altman actually decided that he needed to respond to. Tweeting a few hours later, wrong as you surely know. Want to come visit the first site already underway? This is great for the country. I realize what is great for the country isn't always what's optimal for your companies, but in your new role, I hope you'll mostly put America first.

A critique that has perhaps less competitive complication comes from former congressman and libertarian Justin Amash, who writes, the Stargate project sounds like the stuff of dystopian nightmares. A U.S. government announced partnership of megacorporations, quote, to protect the national security of America and its allies and harness AGI for the benefit of all of humanity. Let's maybe take a beat here.

Now, I don't particularly hold this view, but I do think that Justin is right, at least in the implication that we are going to have to have a large-scale national conversation about where government ends and private enterprise begins in the era of AI. Not that this has ever been an easy line to draw. Obviously, it is one of the key fault lines that divides people when it comes to their political opinions. And there is clearly some lobbying going on around this. On Monday, Scale AI CEO Alexander Wang took out a full-page ad in the Washington Post to

to declare, Dear President Trump, America must win the AI war. In an accompanying open letter, Wang called for the government to emulate the investments made by big tech firms, namely shifting their tech spend to compute. He wrote, China's government outspends our government by about 10 times on AI implementation and adoption. AI progress relies on three pillars, compute algorithms and data. The leading global tech companies have long recognized this by investing approximately 60% on compute, 30% on data, and 10% on algorithms.

This formula is why American commercial AI systems are the best in the world, and why China benchmarks its government spending against us. In contrast, our government spends roughly 90% of investments on algorithms, neglecting the other two pillars. So not only are we spending less, we're not investing well. To win, our government must adapt by changing our investment strategy to more closely reflect industry. Then, we must not only match but exceed China's aggressive funding for AI focused on fielding and implementing AI solutions.

Very short-lived open AI CEO and now founder of a new stealth AI startup, Demit Shear, wrote, This is a horrible framing. We are not at war. We're all in this together, and if we make AI development into a war, we are likely to all die. I can imagine a worse framing, but it takes real effort. Why would you do this?

Wang responded, it's a technology that China is racing to build faster than us so they can use to win militarily and expand their regime. But sure, we can use nice words instead. The conversation goes on for there, perhaps a little bit out of scope for this particular episode, but one we will no doubt come back to, perhaps even for this week's Longreads episode. For now, friends, things are getting real. They are getting big, like a half a trillion dollars big. This will be a very interesting development to watch. For now, appreciate you listening or watching as always. And until next time, peace.

Thank you.