After the Elon Musk fallout, Mark Zuckerberg and Sam Altman move closer to President Trump — but White House officials remain skeptical of the Meta and OpenAI chiefs.

WASHINGTON — It was the seating chart that launched a thousand whispers: Mark Zuckerberg beside President Donald Trump at a White House dinner earlier this month, with OpenAI’s Sam Altman only a few chairs away. The tableau amounted to a high‑definition signal that two of America’s most influential tech leaders are recalibrating their Washington strategy after a very public rift between Trump and Elon Musk. In a capital that reads symbolism like scripture, proximity is policy.

The rapprochement comes after months of volatility in the tech‑politics triangle. Musk — once one of Trump’s most visible allies on technology — fell out with the president over spending plans and regulatory concessions, culminating in a conspicuously empty seat for the Tesla and X owner at a White House dinner that otherwise drew a who’s who of Silicon Valley. Into that vacuum stepped Zuckerberg and Altman, both eager to lock in preferential access as the administration races to turn the United States into the world’s premiere AI arsenal.

For Meta, the incentive is twofold: first, to keep federal antitrust heat from boiling over as the company expands into mixed reality and AI; second, to gain a say in the rules of the road for generative systems that increasingly sit at the center of how billions of people search, shop and socialize. OpenAI’s calculus is different but complementary. The San Francisco research lab turned platform power has been courting Washington with proposals to turbocharge government adoption of AI and to site massive data‑center projects on American soil, often in partnership with cloud and telecom giants. The political logic is simple: build jobs and capacity in swing states; get invited back to the table.

Insiders on both sides say the new alignment is a pragmatic truce rather than a heartfelt conversion. Zuckerberg, who once sparred with Trump over content moderation and civic integrity, has become a more disciplined operator in Washington. His lieutenants now stress economic growth, competitiveness with China and ‘innovation‑first’ guardrails rather than the platform‑policing rhetoric that defined Meta’s posture in earlier election cycles. Altman, for his part, has walked a careful line — publicly evangelizing for safety research and “constitutional” constraints on AI while privately pitching the White House on freedom‑focused policy that accelerates deployment across defense, services and critical infrastructure.

That courtship has coincided with the administration’s expansive AI Action Plan, which seeks to wire federal agencies with large‑language‑model tools and underwrite a once‑in‑a‑generation wave of compute infrastructure. OpenAI‑linked ventures — including mega‑scale data centers with partners in Texas, New Mexico and the Midwest — have been showcased as proof that the plan is already minting jobs. The photo‑ops are potent: groundbreakings with hard hats and rebar, ribbon‑cuttings that promise tens of thousands of local hires, and slide decks featuring gigawatts of new capacity.

Musk’s absence sharpened the contrast. The SpaceX and Tesla chief’s split with Trump deprived the White House of a star who had bridged populist politics and techno‑utopian swagger. But it also reduced the risk of a single figure dominating the AI narrative. With Musk out, aides can elevate a broader cast of executives to validate the administration’s approach — and, at least for now, Zuckerberg and Altman appear willing to play those roles.

Still, senior officials remain wary. The relationship between the administration and Big Tech is laced with muscle memory from past battles: privacy scandals, misinformation blow‑ups, content moderation whiplash and bruising antitrust cases. Several White House staffers describe Meta and OpenAI as useful but untrusted partners — indispensable for capacity and talent, yet structurally conflicted by profit motives and platform power. ‘We need what they build,’ one official puts it, ‘but we also need alternatives in case they blink when the public interest cuts against their quarterly results.’

That skepticism has policy teeth. Procurement officials have quietly pushed for vendor diversity in federal AI pilots, seeding contracts among OpenAI, Google, Anthropic and smaller firms to avoid lock‑in. National security aides want more red‑teaming and model‑eval transparency baked into any government deployment, with clear escalation paths if systems hallucinate, degrade or are exploited. Consumer‑protection lawyers, meanwhile, are pressing for disclosure requirements when citizens interact with AI in public services, from benefits adjudication to immigration interviews.

For Meta, the political risks concentrate around speech. The company eased some fact‑checking and news‑ranking protocols in the name of neutrality and user choice, drawing cheers from conservatives who long accused it of bias — and warnings from civil‑society groups bracing for a turbulent information year. Zuckerberg argues that Meta has matured beyond blunt content interventions and is leaning on labeling, user controls and AI classifiers that throttle the worst abuse at scale. Critics call it capitulation dressed as principle. Either way, the decisions will be tested in the months ahead as campaigns weaponize synthetic media and micro‑targeted ads.

For OpenAI, the flash points are different: safety, scale and sovereignty. Altman has promoted a ‘freedom‑focused’ regulatory vision that rejects heavy licensing but embraces measurable safety standards, third‑party evaluations and incident reporting. The company’s country‑partnership model, pitched as a democratic counterweight to authoritarian AI, is popular in some ministries yet prompts hard questions in Washington: how independent can a vendor be when it is also shepherding national AI strategies? And what happens when its models are upgraded faster than agencies can adapt?

Then there is the market politics. Meta’s hardware‑plus‑AI gambit competes with Apple, Microsoft and a revitalized Google for the next computing interface. OpenAI, while still yoked to Microsoft in the cloud, is increasingly a platform in its own right, with app stores, agents and developer ecosystems that overlap with Big Tech incumbents. For the White House, picking winners is politically perilous; appearing captured by any one company could backfire with voters suspicious of monopoly power. Officials have begun speaking not of ‘a national AI champion’ but of a ‘national capacity’ — the idea that what matters is redundancy and resilience across vendors.

That framing also reflects geopolitical anxieties. U.S. policymakers want to outrun China in advanced chips and AI capabilities without triggering fragile supply chains or a public backlash over corporate subsidies. It is why they talk as much about workforce pipelines and regional development as they do about model sizes. If the administration can trumpet training centers in Ohio as loudly as it does datacenters in Texas, the political math gets easier.

Where does Musk fit now? The answer may be: less as a policy partner, more as an unpredictable foil. His heft in space launch and EVs gives him leverage that no one else can match, but in AI he is now a rival to the very companies that have the president’s ear. That rivalry could prove useful for the White House, which can triangulate among competing egos to extract concessions on safety, disclosure and domestic investment. Or it could devolve into open warfare that slows progress on standards and procurement just as agencies begin to scale deployments.

For Zuckerberg and Altman, the near‑term goal is simple: normalize the photo‑ops and institutionalize the access. That means showing up — at state dinners, ribbon‑cuttings, summits — and delivering tangible wins that policymakers can sell to constituents. It also means swallowing the distrust that still hangs in the air. In Washington, everyone likes a friend who can build things. Fewer trust a friend who also owns the rails.

The tech bromance, in other words, is real — but it’s also a marriage of convenience, bounded by leverage and watched by skeptics. If it delivers jobs, safer systems and a sturdier public sector, the doubts may fade. If not, expect the seating chart to change again, and soon.