President Donald Trump’s administration faces a growing internal divide over artificial intelligence regulation. Advisers urge caution over security risks while David Sacks and major tech executives advocate for a lighter regulatory approach.

White House Officials Raise AI Security Concerns

The disagreement intensified in May after Sacks persuaded Trump to abandon a planned executive order that would have given the government a longer window to review powerful AI models, the Wall Street Journal reported on Wednesday.

White House Chief of Staff Susie Wiles and Treasury Secretary Scott Bessent were reportedly caught off guard by the decision. Bessent has raised concerns that AI-powered cyberattacks could disrupt the nation’s banking system.

At Wiles and Bessent’s urging, Trump later signed a scaled-back version of the order. Since then, Wiles, Bessent and other officials have reportedly held frequent discussions about AI risks and potential safeguards. White House officials have also spoken with executives at OpenAI and Anthropic about their concerns.

Mythos, AI Agents Raise Alarm

Anthropic’s Mythos model reportedly became a major warning sign for officials after they recognized its potential to facilitate sophisticated cyberattacks and create national-security risks. A senior administration official described the model as a "red flashing light."

Concerns grew further after an OpenAI incident in which hundreds of AI agents acted in coordination during a hack of Hugging Face. Officials have since examined potential threats involving critical infrastructure, including water and power systems, as well as the possible use of AI in biological weapons development.

Sacks, Tech CEOs Push Back on Regulation

Sacks has continued advocating limited government intervention, saying existing product-liability and fraud laws are largely sufficient for AI models. "It is the responsibility of these companies to maintain control of their products," Sacks said Wednesday at a Politico event.

Nvidia Corp (NASDAQ: NVDA) CEO Jensen Huang and Meta Platforms, Inc. (NASDAQ: META) CEO Mark Zuckerberg have pushed back against calls for coordinated AI slowdowns and additional regulation. Tesla Inc. (NASDAQ: TSLA) and Space Exploration Technologies Corp. (NASDAQ: SPCX) CEO Elon Musk, meanwhile, has supported calls for greater AI safeguards.

Trump has maintained that the U.S. must accelerate AI development to compete with China. On Truth Social, he wrote: "WHOEVER WINS AI, WINS!"

White House spokeswoman Liz Huston told the publication that the administration remains committed to "unleashing American innovation" while supporting the safe deployment of advanced AI and protecting national security. The White House did not immediately respond to Benzinga’s request for comment.

China Warns AI Could Threaten Political Stability

Meanwhile, Chen Yixin, China’s minister of state security, has warned that AI could pose a direct threat to the Chinese Communist Party’s hold on power. Chen’s concerns reportedly include deepfakes and AI-generated content that could disrupt political stability, as well as the potential exposure of sensitive information through foreign AI models.

Former Vice President Kamala Harris and former President Barack Obama have also voiced support for slowing the pace of AI development.

Trump is set to hold talks on Iran with Gulf leaders next week, according to a report by Axios.

Diplomatic engagement on Iran

The planned discussions bring together Trump and Gulf leaders to address matters related to Iran. The report, attributed to Axios, indicates the meetings are scheduled for next week, though specific dates, locations, and agenda details were not provided in the available information.