The Trump administration is considering a plan to create an independent regulator modeled on Wall Street’s brokerage watchdog to vet the safety of advanced artificial intelligence models, according to people familiar with the proposal. The plan, developed with input from Treasury Secretary Scott Bessent, is now under review by White House Chief of Staff Susie Wiles and comes as the administration seeks to balance rapid AI innovation with growing industry calls for regulatory clarity. The move marks a notable evolution in how Republicans are approaching AI governance after months of a lighter-touch, voluntary framework.
Story Highlights
- The proposed regulator would be modeled on FINRA, the industry-funded watchdog that oversees brokerage firms under SEC authority
- Treasury Secretary Scott Bessent helped develop the plan, which remains under review by Chief of Staff Susie Wiles
- The push follows complaints from Silicon Valley leaders about inconsistent federal action and comes after a new Chinese AI model rattled markets
What Happened
According to reporting from Bloomberg, the Trump administration is weighing a proposal to establish an independent regulatory body that would oversee the safety of advanced artificial intelligence models, with structured input from the AI industry itself. The plan was developed in part by Treasury Secretary Scott Bessent and would create an agency reporting to the Securities and Exchange Commission, similar in structure to the Financial Industry Regulatory Authority, the self-regulatory body that currently oversees brokerage firms. The proposal is currently being reviewed by White House Chief of Staff Susie Wiles, whose office has generally involved itself in major policy debates, such as tariffs or AI regulation, only when the political stakes are significant enough to require coordination across multiple agencies.
The idea has been publicly championed by prominent AI executives, including Demis Hassabis, who compared the proposed watchdog to FINRA in a post on social media and has scheduled meetings with Washington policymakers to advocate for the framework. Sources say the plan aims to resolve friction between two constituencies that have each grown frustrated with the administration’s current approach: Wall Street firms, who are concerned about the cybersecurity risks posed by increasingly powerful AI systems, and Silicon Valley developers, who say the government’s ad hoc interventions, including a previous request that OpenAI modify its Sol model, have created uncertainty without a clear, predictable process.
The proposal comes on the heels of a very different approach the administration took just weeks earlier. On June 2, President Donald Trump signed an executive order titled “Promoting Advanced Artificial Intelligence Innovation and Security,” which established a voluntary framework allowing AI companies to submit their most advanced models for government testing before public release. That order explicitly prohibited the creation of new mandatory licensing, preclearance, or permitting requirements, reflecting a deregulatory instinct shaped in part by former White House AI and Crypto Advisor David Sacks. The new FINRA-style proposal would represent a meaningful departure from that voluntary model toward a more structured, though still industry-influenced, oversight regime.
Trump himself has not yet reviewed the new plan, according to one source familiar with the matter, though officials have accelerated efforts to bring greater clarity to AI policy following the release of a new Chinese AI model last week that triggered a notable drop in the stock prices of companies linked to the sector. Many details of the proposal remain unresolved, including what specific model evaluations would be required, how a new agency would be funded, and what precise role the SEC would play in overseeing AI developers going forward.
Why It Matters
For Republicans who have championed a deregulatory approach to emerging technology, the consideration of a new independent AI regulator represents a meaningful policy crossroads. The administration has spent much of the past year pushing back against a patchwork of state-level AI laws, including through a Department of Justice AI Litigation Task Force established in December 2025, on the theory that federal consistency serves innovation better than fragmented state rules. A federal regulator, even an industry-informed one modeled on FINRA, would mark a shift from pure deregulation toward a more formalized, if still light-touch, oversight structure.
For American technology companies, particularly leading AI labs such as OpenAI and Anthropic, both of which are reportedly considering initial public offerings in the coming year, regulatory clarity carries direct commercial significance. A predictable, industry-informed review process could provide the kind of certainty needed to support public listings and long-term capital investment, addressing concerns that ad hoc government interventions have created unpredictable business risk.
For national security policymakers, the timing tied to a competitive Chinese AI model underscores the broader strategic stakes. Ensuring that the United States maintains its technological edge while also addressing legitimate safety and cybersecurity concerns has become a central preoccupation of the administration’s broader AI strategy, one that intersects directly with concerns about Beijing’s rapid advances in the sector.
Economic and Global Context
The release of a competitive Chinese AI model last week triggered a real market reaction, with shares of companies tied to the AI sector declining amid renewed anxiety about America’s technological lead. That episode appears to have accelerated internal administration deliberations, according to sources familiar with the timeline, illustrating how closely intertwined domestic AI policy has become with broader U.S.-China strategic competition.
The proposed FINRA-style model would be industry-funded, similar to its financial sector counterpart, potentially minimizing direct costs to taxpayers while still providing a structured evaluation mechanism for advanced models. This funding structure reflects a broader Republican preference for market-based and industry-financed oversight mechanisms over traditional government bureaucracies funded through general appropriations.
Globally, the world’s major economies remain in various stages of developing their own AI governance frameworks, from the European Union’s comprehensive regulatory approach to lighter-touch models favored by the United States. How Washington ultimately structures its oversight regime will likely influence international norms and could shape competitive dynamics between American, Chinese, and European AI developers for years to come.
Implications
In the near term, expect continued internal deliberation within the White House as officials weigh the tradeoffs between the existing voluntary framework and the new proposed regulator. Given that Trump himself has not yet reviewed the plan, its ultimate fate remains genuinely uncertain, and the final structure, if adopted, could differ substantially from what is currently under consideration.
For AI companies, particularly those weighing public offerings, the coming months will be critical in determining what kind of regulatory environment they will operate within as they approach capital markets. Clarity on evaluation requirements and funding structures will likely factor directly into IPO timelines and investor confidence.
For congressional Republicans, the proposal offers an opportunity to shape legislation that could codify aspects of the new framework, particularly if the administration ultimately decides that formal congressional action is needed to establish the regulator’s authority. Given the bipartisan interest in AI safety, there may be room for cooperation, even as broader disagreements over the scope of AI regulation persist between the parties.
Sources
