A person in a suit speaks at a podium with a microphone, addressing a crowd outside the US Capitol.

Chris Van Hollen Says AI's Rapid Growth Is 'Extremely Dangerous for Humanity,' Urges Lawmakers to Ban Recursive Self-Improvement

Sen. Chris Van Hollen (D-Md.) has warned that rapid advances in artificial intelligence (AI) pose growing risks to humanity, urging Congress to introduce immediate safeguards, hold AI companies accountable, and consider banning systems that can independently improve themselves.

Van Hollen Demands Urgent AI Regulations

On Wednesday, in a post on X, Van Hollen argued that AI’s rapid development was benefiting wealthy investors while exposing ordinary Americans to significant risks.

“Anthropic’s $1 trillion IPO & OpenAI’s newest model highlight how AI’s rapid development is great for the billionaire class but extremely dangerous for humanity,” he wrote.

In an accompanying video, Van Hollen said recent AI developments showed “how urgent it is to address the risks posed by AI.”

He called for “mandatory and independent third-party evaluation” of AI companies and said they should be held “accountable, liable, and responsible for any harm they do.”

The senator also called for protections for American workers, warning about job disruptions, privacy violations and rising electricity bills.

“We need to make sure that all of you aren’t paying higher electric bills for data centers built by the biggest and richest corporations on the planet,” he said.

Senator Calls For Ban On AI Self-Improvement

Van Hollen also raised concerns about AI agents escaping containment, facilitating bioweapons development and enabling cyberattacks.

Most notably, he urged lawmakers to consider banning recursive self-improvement, in which AI systems help train and develop increasingly advanced versions of themselves.

“The fact that AI begins to train AI and get beyond human capacity to intervene. That could put us all at risk,” he warned.

He added, “We need to move quickly to address that big threat to all of us.”

New York Introduced Stricter AI Safety Rules

Last week, New York launched a registration portal requiring major AI developers, including OpenAI and Anthropic, to disclose safety practices under a new law taking effect Jan. 1.

The law mandated safety reports, disclosure of critical incidents within 72 hours and penalties of up to $3 million for repeat violations.

New York Gov. Kathy Hochul criticized companies for prioritizing AI development over public safety, saying, “Any collateral damage is worth it as long as we win the race to the AI cliff.”

AI Safety Push Gains Momentum

On Tuesday, Sen. Mark Kelly (D-Ariz.) prepared legislation for stricter AI oversight, calling for “real rules, real oversight and real consequences” for companies violating safety requirements.

Previously, Pentagon technology chief Emil Michael said the administration planned to outline existing AI regulations, including third-party audits and potential export restrictions for violations.

Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.

Image via Shutterstock

Market News and Data brought to you by Benzinga APIs
Share this article