Trump's New AI Dream Team: Zuck, Huang, Ellison

Trump's new AI dream team, Zuck, Huang, Ellison, three of the most powerful men in tech just got formally pulled into the White House. Mark Zuckerberg, Jensen Huang, Seri, Ellison. As of March the 25th, 2026, all three are sitting on Trump's new AI advisory council. It's called PCAST, the President's Council of Advisors on Science and Technology.

And the people on this council are about to have a direct say in how AI gets built, regulated and deployed across America. That matters for you, whether you run a business, a content operation, an agency or a store, because the decisions made in that room will shape what AI tools exist, how fast they develop and what rules govern them. Let me break down exactly who's in the room, what they're working on and what it means for you. So first, what is PCAST?

It's a Presidential Science Advisory Council. This existed since 1933. Every president since FDR has had one. The idea is simple, bring in outside experts to help the White House make better decisions on science and technology.

Obama's version have mostly academics and researchers. Biden's had 27 members from universities, non-profits and industry, deliberately broad. Trump's version has 13 members and this one is built very, very differently. 12 of the 13 are CEOs, investors or founders of major tech companies.

One is a scientist, John Martinis, a Nobel Prize winning quantum physicist from UC Santa Barbara. The full list, Mark Anderson from A16Z, Sergey Brin, Google co-founder, Safra Katz and Larry Ellison from Oracle, Michael Dell from Dell Technologies, Fred Ersom from Coinbase, Jensen Huang from NVIDIA, Lisa Su from AMD, Mark Zuckerberg from Meta. Plus two energy founders, Jacob DeWitt from Nuclear Startup Oklo and Bob Mungard from Commonwealth Fusion Systems and David Freeburg, a venture investor. This is the most industry heavy version of PCAST in the council's history.

The administration's bet is clear, if you want to win the AI race, you bring in the people who are actually building it. Now, let's talk about why each of the big three matters. Start with Jensen Huang. NVIDIA just reported $215 billion in revenue for Fiscal 2026, up 65% year on year.

Their data center business, the chips that power AI, brought in $193.7 billion of that alone. Every AI tool that you use right now almost certainly runs on NVIDIA hardware. Chachibity, Claude, Midjourney, Gemini, the chips inside those systems are overwhelmingly NVIDIA. Huang's seat on PCAST puts the person who builds the backbone of global AI infrastructure, directly in conversation with the White House on policy.

The big question his presence will shape, chip export rules. How much of this hardware gets sold to whom, at what price, under what terms. Those decisions affect the pace of AI development globally, and the cost of AI tools for everyone downstream. Chinese competitors and companies have already placed orders for over 2 million of NVIDIA's most powerful chips.

That's up to $14 billion in potential sales. What happens with those orders, and dozens of decisions like them, will be influenced by the framework that this council helps shape. Now Zuckerberg. Meta is spending between $115 and $135 billion on AI infrastructure in 2026.

That's more than any company has ever spent on technology in a single year. Their 2025 spend was $64 to $72 billion. They're doubling it. When Zuckerberg was appointed he said, the United States has the opportunity to lead the world in AI.

I'm honored to join the president's council and work with other industry leaders to help make this happen. But what does Meta actually want out of this council? Three things. First, rules that support open-source AI.

Meta's LLAMA models are the dominant open-source AI ecosystem right now. A regulatory environment favorable to open-source means Meta's approach wins globally. Second, federal standards that simplify the regulatory landscape. Right now there are over a thousand state-level AI bills that were introduced in 2025.

118 of them were enacted across 38 states. For a company like Meta operating nationally, that's a compliance nightmare. A single federal standard fixes that. And third is speed.

Meta is betting enormous capital on AI being the next platform. And the faster the regulatory environment allows them to build and deploy, the bigger the return on that bet. For you as a business owner, if Meta's open-source push wins, that means more powerful free AI tools. LLAMA-based models already power a huge chunk of the AI products smaller businesses use.

More support for open-source means more competition, more options, and therefore lower costs. Now Ellison. Oracle is one of the less glamorous names on this list, but arguably the most important for understanding where AI infrastructure is actually going. Oracle Cloud Infrastructure, OCI, is becoming one of the primary places where AI workloads actually run.

They have a reported $300 billion contract with OpenAI. Their remaining performance obligations, basically future contracted revenue, hit $523 billion. Their cloud backlog is growing faster than AWS and Azure. Larry Ellison was part of the Stargate announcement in January 2025, the $500 billion AI infrastructure initiative that kicked off this whole era of massive AI investment.

Ellison's presence on the council means the person building the cloud infrastructure that AI runs on is directly advising on data policy, energy permitting, and the federal contracts that flow to cloud providers. The administration already issued an executive order last year, fast-tracking permits for data centers over 100 megawatts on federal land. More of that direction is where this council will push. Here's where I want to zoom out and tell you what this council is actually working on, because this is where it gets relevant to your business.

Five days before the PCAST appointments were announced, the White House released a national AI legislative framework. Four main positions. Number one, training AI models on copyright material does not violate copyright law, according to them. This is significant for anyone in creative industries.

It means the administration's position is that AI companies can train and publish content. Number two, federal rules should override state-level AI laws. One national standard instead of 50 different state rules. Number three, no new AI regulatory body.

Existing agencies can handle it. And number four, regulatory sandboxes, right? So companies can experiment with AI under relaxed rules before those rules are formally set. PCAST doesn't write laws, it produces recommendations.

But the two people running PCAST, David Sachs, Trump's AISR, and Michael Krasios, the head of the White House Office of Science and Technology Policy, are the two most powerful tech policy officials in the executive branch. The council's recommendations land directly on their desks. That's the pipeline. Industry advice flows into Sachs and Krasios, who shape executive orders and legislative proposals which set the rules everyone else operates under.

Quick one, if you want to stay exactly on top of all this sort of stuff, if you want to learn what to do when this sort of development starts shaping business, and you want to know how to save time, get more customers, and grow with AI automation, check out the AI Profit Boarding. Link in the comments description or go to the AIProfitBoarding.com. Now here's a part of the story that I think is genuinely interesting that nobody's really covering. Look at who's not on the council.

Elon Musk isn't there. His relationship with Trump broke down publicly in mid 2025. Trump said Musk had lost his mind. Musk formed his own political party.

That's done. Sam Altman isn't there. Even though Altman literally stood at the White House in January 2025 for the Stargate announcement, the biggest AI infrastructure deal in history, he's not on PCAST either. So who is there?

Well, Huang, Ellison, Dell, Su, Bryn, these are all infrastructure and hardware people. They build the physical layers of AI, the data centers, the chips, the cloud, not the model builders, not the chatbot companies, the actual pipes. That tells you exactly where the Trump administration thinks the real game in AI is being played. It's not the software on top.

It's the infrastructure underneath. Whoever controls the compute, the chips, the power, the data centers, controls the pace and direction of AI deployment globally. That's a bet this council reflects, and that framing matters for how you think about AI tools in your business. The tools will keep getting cheaper and better, but the rate at which that happens and which companies are best positioned to deliver them is being shaped right now by these infrastructure decisions.

There's one more thing worth noting about the timing of all this. Trump's first term had a science advisory council too. It took 33 months to appoint anyone to it. It produced one report.

This time the council was established by executive order on day one, January the 23rd 2025. It still took 14 months to fill the seats, but when they filled them, they filled them with the people spending hundreds of billions of dollars on AI right now. The signal is clear. This administration sees AI as a defining technology competition of the next decade, primarily against China.

And it wants the people building American AI infrastructure to have a direct voice in how that competition is run. The practical reality is this. The rules governing AI in America are being set right now. The framework is being written and the people writing it are the people whose companies your AI tools are built on top of.

That's not a reason to panic. It's a reason to pay attention because the business owners who thrive through this period won't be the ones who were surprised by how it went. They'll be the ones who saw it coming and adopted and adapted early. The AI tools used today, the ones used to write content, manage customers, run ads, automate your workflows, their price, their capability, and the rules around how you can use them are all downstream of decisions being made in rooms like this one.

Understanding who's in those rooms is the first step to staying ahead. And that's exactly what I want to help you do. If you're an agency owner, a freelancer, an e-commerce operator, and you want weekly coaching calls, daily tutorials, and a community of 2,600 business owners who are focused on helping you save time, grow, and get more customers, then feel free to get that in the AI Profit Boardroom. Link in the comment description or go to the AIProfitBoardroom.com.

That's Trump's new AI dream team, Zuck, Huang, Ellison. The people building the future of AI infrastructure just got a direct line to the White House. Now you know what they're working on, and I'll see you in the next one.