
AI adviser Krishnan says no federal licensing for AI. But 75 data center projects halted by local opposition show state pushback is the real regulator. Market implications.
The White House will not create a formal licensing regime for artificial intelligence, outgoing AI adviser Sriram Krishnan said in an interview with the Financial Times. “There will not be an FDA for AI,” Krishnan said, calling centralized oversight “burdensome, onerous, bureaucratic red tape.” Establishing a federal agency that required “a team of lawyers before you can get a model out” would put “sand in the gears” of progress, he added. “That is never, never going to happen under President Trump.”
The administration’s hands-off stance sits alongside a track record of ad hoc intervention. Weeks before Krishnan’s interview, the federal government forced Anthropic to pull the latest version of its Mythos model and paused the launch of OpenAI’s 5.6 – moves the FT called “unprecedented.” Krishnan, a former venture capitalist who worked with Elon Musk, has been one of the strongest voices inside the White House for a laissez-faire approach.
Yet the absence of a federal licensing regime does not mean AI companies face a free pass. Local opposition is filling the regulatory vacuum. Researcher Data Center Watch tallied 75 data center projects halted by community pushback in the first three months of the year alone. A majority of Americans now support tough AI regulations, the same firm found. The disconnect between Washington’s policy and public sentiment is widening.
The market read: the binding constraint on AI infrastructure is no longer federal rulemaking. It is town councils, utility boards, and zoning hearings. Without a federal body to override local decisions, every new data center or compute cluster becomes a test of local political will. That shifts risk from compliance costs to project delays and legal uncertainty. A buildout that clears one municipality may hit a wall in the next.
Meanwhile, companies themselves are drawing a line before autonomous execution. PYMNTS Intelligence research shows that 100% of wholesale firms surveyed restrict AI agents to “look-up access” – the agent can retrieve information but cannot act on it. The figure stood at 90% for retailers and 85% for construction companies. No firm in any sub-industry fully permits autonomous AI action. “The restriction is not a sign that AI deployment has stalled,” PYMNTS wrote. “The picture is of companies that have embedded AI broadly for research, analysis and document generation, and drawn a line before execution.”
The combined picture: a White House that refuses to centralize control, a public that wants more oversight, and an industry that self-limits its most powerful tools. For investors, the next concrete marker is the Q2 data center project tally from Data Center Watch, expected in August. Until then, the regulatory landscape will be written county by county.
Prepared with AlphaScala research tooling and grounded in primary market data: live prices, fundamentals, SEC filings, hedge-fund holdings, and insider activity. Each story is checked against AlphaScala publishing rules before release. Educational coverage, not personalized advice.