The Trump administration is exploring whether to introduce targeted federal oversight of cutting-edge artificial intelligence systems, a notable evolution from its earlier light-touch approach to the fast-growing industry.

According to U.S. officials and individuals familiar with internal discussions who spoke with The New York Times, the White House has been considering an executive order that would establish a joint working group of government officials and technology executives.

The panel would reportedly evaluate whether new safeguards — including a potential review process for advanced AI models — are needed.

The deliberations come as President Donald Trump continues to promote rapid AI development as key to U.S. competitiveness, particularly against China, while signaling caution about overregulation.

"We're going to make this industry absolutely the top, because right now it's a beautiful baby that's born," Trump said in July. "We have to grow that baby and let that baby thrive."

"We can't stop it. We can't stop it with politics. We can't stop it with foolish rules and even stupid rules," he added.

At the same time, Trump indicated any guardrails must be carefully crafted, saying they "have to be more brilliant than even the technology itself."

The internal reassessment follows the debut of Mythos, a powerful AI system developed by Anthropic that can detect software vulnerabilities at an advanced level. The company has withheld public release of the model, warning it could spark a cybersecurity "reckoning."

Officials are increasingly focused on the dual-use nature of such systems — both the risks of malicious cyberattacks and the potential benefits for U.S. military and intelligence operations.

Some within the administration support a framework that would allow federal agencies early access to new models without preventing companies from bringing them to market.

"The technology is moving extremely fast, and there are few formal procedures, but they also don't want to overregulate," Dean Ball, a former Trump administration AI adviser, told the Times. "It's a tricky balance."

A White House official downplayed reports of imminent action, telling Newsmax that talk of an executive order was "speculation" and emphasizing that Trump would personally announce any decisions.

The policy discussions are unfolding amid leadership changes, with Chief of Staff Susie Wiles and Treasury Secretary Scott Bessent taking on larger roles after the departure of former AI czar David Sacks.

Complicating matters is an ongoing dispute between Anthropic and the Pentagon over a $200 million contract tied to military AI use. The disagreement led the War Department to halt use of the company's tools earlier this year, prompting a lawsuit from Anthropic.

Despite the conflict, the company's technology remains embedded in parts of the national security apparatus. Its systems have reportedly supported intelligence analysis in military operations and have been used by the National Security Agency to probe government software for vulnerabilities.

Wiles and Bessent recently met with Anthropic CEO Dario Amodei in an effort to restore broader cooperation, with both sides later describing the talks as "productive."

If the administration moves forward, multiple agencies — including the NSA and top intelligence offices — could play a role in shaping or carrying out any review process.

Officials may also revisit the role of an AI standards body created under the Biden administration but largely sidelined since Trump returned to office.

Newsmax reached out to the White House for comment but did not receive an immediate response.

• Trump to Newsmax: AI Will Lead to 'Tremendous Numbers of Jobs'
• State Dept Warns of Alleged AI Thefts by DeepSeek, Other China Firms