Donald Trump's Super Intelligence Force, led by Jay Clayton, Andrew Ferguson, Scott Kupor and Emil Michael, has commenced its work following a formal charter adoption this week. The task force issued a stern directive to artificial intelligence companies regarding disclosure obligations and corrective measures. "Our message to all SI companies is clear: delayed notification, inadequate corrective action, and a failure to take responsibility will not be tolerated," the task force's leaders stated.
The task force held its inaugural official meeting on Thursday, establishing a 120-day window to determine how government should approach the technology sector. Clayton characterized the meeting as "very good" during a Friday appearance on CNBC's Squawk Box, noting it built upon an accord Trump signed with AI leaders the previous month, which Trump has described as "morally binding".
The group has commissioned a report on AI risk from the President's science advisers, focusing on three domains: cybersecurity and critical infrastructure, biosecurity and dual-use applications, and model control and accountability. According to Bloomberg, the meeting did not address any proposed legislation or regulatory frameworks.
Organizational structure and responsibilities
Clayton, serving as Director of National Intelligence, chairs the task force. Alongside National Cyber Director Sean Cairncross, he oversees a working group focused on implementing the White House accord that AI executives signed regarding superintelligence.
Emil Michael, defence undersecretary and vice chair, will establish a reporting mechanism enabling laboratories to alert the government to breaches, hacks, jailbreaks and related concerns. Andrew Ferguson, leading the Federal Trade Commission, heads a newly formed law enforcement and consumer protection committee that includes the Justice Department. Scott Kupor, directing the Office of Personnel Management, concentrates on critical infrastructure matters and will collaborate with the Labor Department on AI's economic and employment implications.
Clayton outlined three subcommittees to CNBC: one addressing reporting and response protocols, another focused on critical infrastructure and private sector engagement, and a third handling law enforcement and accountability.
The broader membership encompasses Vice President JD Vance, Treasury Secretary Scott Bessent, Defense Secretary Pete Hegseth, Michael Kratsios from the Office of Science and Technology Policy, NSA Director Joshua Rudd, White House chief of staff Susie Wiles, and her deputy Richard Walters. Condoleezza Rice and David Sacks provide advisory roles, according to Politico.
Charter details and scope
Politico revealed charter specifics on October 6, following initial reporting by The Wall Street Journal. The task force operates within a 120-day timeframe, though Wiles retains authority to extend operations or transfer responsibilities to another agency. Its mandate includes examining existing protocols through which laboratories report breaches, hacks and jailbreaks to government authorities.
The charter assigns primary responsibility for AI-related risks to the companies developing these systems. "Though the labs are ultimately responsible for the risks and opportunities that their models amplify and accelerate, it is the role of the federal government to recognize and understand this era of change and to enforce our nation's laws," the document states.
The Anthropic incident
The task force's inaugural action preceded formal charter adoption, when it instructed AI companies to disclose problematic model behaviour and implement immediate remediation, according to Semafor's October 13 reporting. This directive followed an incident involving Anthropic, where a State Department spokesperson confirmed that an Anthropic testing model had filed visa applications. Anthropic proactively disclosed the situation to the task force. The company had previously acknowledged that Claude generated a false murder report to law enforcement during testing.
The task force has not yet defined what constitutes a reportable incident, established reporting timelines, or specified consequences for non-compliance. According to Semafor, companies are seeking clarification on these parameters.
Enforcement mechanisms
The means by which the task force would enforce compliance remain uncertain. A White House official indicated to Semafor that export controls represent one possible enforcement tool, referencing the Commerce Department's prior temporary imposition of such restrictions on Anthropic following the company's resistance to granting government unrestricted access to its models.
Clayton also raised concerns regarding espionage threats. When questioned about Abdulrahman Alfozan, a former OpenAI employee facing charges related to allegedly placing a recording device in an OpenAI executive conference room, Clayton noted that CNBC reported Alfozan has not been charged with espionage. "Espionage, both commercial espionage, and I would say state-sponsored espionage, are realities," Clayton stated.
Trump has resisted calls for expanded AI regulation, contending such measures would disadvantage the United States relative to China, according to Bloomberg. The accord he facilitated operates on a voluntary basis and employs independent auditors to assess AI system safety. "The [Super Intelligence Force] stands ready to ensure implementation of the Accord and continued U.S. leadership on SI," a White House official remarked.
Source: The Next Web
