Jay Clayton's AI task force cannot enforce real safety rules

By Nikhil Raghavan · Reporting from San Francisco · October 4, 2026

Why a Wall Street regulator is keeping watch on superintelligence The White House has a new name for its artificial intelligence policy. It sounds less like policy and more like a comic book title. President Donald Trump tapped Director of National Intelligence Jay Clayton to lead the newly formed "Super Intelligence Force" (SI), according to CNBC and Anadolu Agency. The task force has 120 days to research risks and opportunities. It will offer recommendations on federal responsibilities. It arrives after a White House meeting where tech executives signed a "morally binding" agreement on safety standards, as reported by CBS News.

Jay Clayton's selection reveals the administration's actual appetite for binding rules. Clayton previously chaired the Securities and Exchange Commission. He also served as U.S. attorney for the Southern District of New York. He took over the Office of the Director of National Intelligence in July, giving him authority over 18 U.S. intelligence agencies, as noted by CNA and Der Spiegel. He spent two decades at Sullivan & Cromwell representing Wall Street banks. During his SEC tenure, corporate penalties and insider trading actions fell while the agency favored disclosure over mandates.

Now he must ensure America leads in superintelligence without hindering growth. This setup shares the mechanism of the Volcker Rule, section 619 of the Dodd–Frank Act. Proposed in 2010 to restrict banks from speculative investments, that rule banned proprietary trading while carving out specific exceptions. Clayton’s task force similarly favors soft compliance over hard prohibitions. It expects that "dass im Dialog Mechanismen entwickelt werden, so wie es in anderen Branchen der Fall ist," as Der Spiegel reported.

The friction between voluntary promises and intelligence oversight Industry leaders offer the strongest case against this light-touch approach. Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman called for federal guardrails. Altman signed statements warning that mitigating extinction risks from artificial intelligence should be a global priority alongside pandemics and nuclear war. These executives argue that generative models drop the cost of generating believable misinformation by several orders of magnitude.

Yet this anxiety collides with the administration's broader deregulatory posture. President Trump dismissed AI fears as a hoax and blamed "Radical Left Dumocrats." He argued that slowing domestic development hands an edge to Chinese labs. To reconcile these views, the White House convened tech leaders to sign internal control commitments. Trump described these promises as "almost like a constitution." Voluntary standards enforced through self-policing are not a regulatory framework. They are an armistice signed by firms that profit directly from the technology.

Who answers the page when the model breaks To see how this ends, ask who gets paged at three in the morning when a model fails. It will not be David Sacks, who stepped down as AI and crypto czar in March. Nor will it be venture capitalists who dismiss warnings. It will be intelligence community engineers reporting to an agency head wearing two hats. They must reconcile corporate trade secrets with national security threat-sharing mandates.

The 120-day clock ticking down for Jay Clayton's Super Intelligence Force will not yield statutory guardrails. It will produce a disclosure-heavy report. That report will protect market leaders, satisfy intelligence monitoring, and leave liability to Congress. When the next major exploit hits, a voluntary West Wing agreement will not patch the code.

Jay Clayton's AI task force cannot enforce real safety rules

Sources