Kevin Hassett, the Director of the White House National Economic Council, is pushing back against proposals that would create a sprawling new government apparatus to evaluate and approve artificial intelligence models. His position, articulated on May 6, 2026, lands squarely in the middle of an intensifying debate over how Washington should handle AI systems that are growing more powerful by the quarter.
The FDA model, minus the FDA
The administration is reportedly considering an executive order that would create evaluations for AI models similar to how the Food and Drug Administration vets new drugs before they hit the market. The concept surfaced around May 7, 2026, and it immediately raised a question that Hassett seems eager to answer: does this mean we’re building an FDA for algorithms?
His answer is a firm no. Rather than stand up a brand-new federal body, Hassett has argued for leveraging existing government frameworks to conduct these assessments.
Anthropic disclosed an advanced AI model called “Mythos” in early May 2026, and the capabilities on display apparently accelerated security-focused conversations inside the White House.
The China argument and the 4% GDP bet
Hassett isn’t just making a process argument. He’s making a geopolitical one. His concern, echoed in broader policy discussions highlighted around May 11, 2026, is that excessive regulation could drive AI innovation overseas, specifically to China.
Hassett has paired this warning with a notably optimistic economic forecast. On May 6, 2026, he projected 4% US GDP growth, attributing the anticipated expansion primarily to enhanced AI productivity and increased capital investments. For context, the US economy hasn’t sustained 4% annual growth in decades. That kind of number would represent a significant departure from the roughly 2-3% range that has defined recent history.
What this means for investors and the crypto-AI intersection
For tech investors, Hassett’s stance is a signal worth tracking closely. A regulatory environment that avoids building a new approval bottleneck could meaningfully benefit companies developing and deploying AI at scale.
If the executive order lands closer to Hassett’s vision, using existing agency structures rather than creating new ones, it would likely be received as a positive by markets that have been watching Washington’s AI posture with increasing anxiety.
For the crypto sector, the connection is less immediate but still worth watching. AI and blockchain technologies are increasingly converging, particularly in areas like decentralized computing networks that offer GPU resources for AI training and inference.
The risk to monitor is scope creep. Even if the initial framework avoids creating a new agency, the evaluation criteria themselves could become a de facto gatekeeping mechanism if they’re designed poorly.
Investors in both AI and crypto-AI projects should be watching the specific language of any forthcoming executive order, not just the headline framing, to understand what “existing frameworks” actually means in practice and which agencies end up holding the keys.
