Anthropic Drops Hallmark Safety Pledge in Race With AI Peers

Anthropic, which for years billed itself as a safer alternative to artificial intelligence rivals, has loosened its commitment to maintaining its guardrails, one of the most dramatic policy shifts in the AI industry yet as startups once focused on helping humanity turn their attention to profit and success. The company in 2023 said in its Responsible Scaling Policy that it would delay AI development that might be dangerous. In a Tuesday blog post, Anthropic said it was updating its rules to say it would no longer do so if it believes it lacks a significant lead over a competitor. “The policy environment has shifted toward prioritizing AI competitiveness and economic growth, while safety-oriented discussions have yet to gain meaningful traction at the federal level,” Anthropic said in its post. Bloomberg News Tech & National Security Reporter Katrina Manson joins Bloomberg Businessweek Daily to discuss. She speaks with Carol Massar and Tim Stenovec. Anthropic PBC, which for years billed itself as a safer alternative to artificial intelligence rivals, has loosened its commitment to maintaining its guardrails, one of the most dramatic policy shifts in the AI industry yet as startups once focused on helping humanity turn their attention to profit and success.

The company in 2023 said in its Responsible Scaling Policy that it would delay AI development that might be dangerous. In a Tuesday blog post, Anthropic said it was updating its rules to say it would no longer do so if it believes it lacks a significant lead over a competitor.

“The policy environment has shifted toward prioritizing AI competitiveness and economic growth, while safety-oriented discussions have yet to gain meaningful traction at the federal level,” Anthropic said in its post.

The company’s move underscores how the high-minded intentions that guided AI startups in their early years have increasingly collided with the pressures to make money and beat out the competition. Anthropic is racing for dominance in the revolutionary technology against a host of formidable rivals, including OpenAI, Alphabet Inc.’s Google and Elon Musk’s xAI Corp.

Dario Amodei, chief executive officer of Anthropic, used to work at OpenAI and left in 2020 in part because of his concerns that the startup was prioritizing commercialization and speed over safety.

OpenAI began as a nonprofit and converted to a more traditional for-profit enterprise last year. It also updated its mission statement in 2024 by dropping the word “safely” from its goal of ensuring that artificial general intelligence benefits humanity.

Both companies are now pushing for initial public offerings as soon as this year, seeking to tap investor interest in AI. Anthropic was recently valued at $380 billion, while OpenAI is raising money at a valuation of more than $850 billion. “From the beginning, we’ve said the pace of AI and uncertainties in the field would require us to rapidly iterate and improve the policy,” an Anthropic spokeswoman said.

The updated policy coincides with a growing dispute with the US Defense Department over Anthropic’s insistence on guardrails for use of its Claude AI tool. The Pentagon on Tuesday threatened to invoke a Cold War-era law to compel Anthropic to allow the US military to use the startup’s technology if the company failed to comply with the government’s terms by Friday.

--------

Watch Bloomberg Radio LIVE on YouTube

Weekdays 7am-6pm ET

Follow us on X

Subscribe to our Podcasts:

Bloomberg Daybreak

Bloomberg Surveillance

Bloomberg Intelligence

Balance of Power

Bloomberg Businessweek

Listen on Apple CarPlay and Android Auto with the Bloomberg Business app:

Apple CarPlay

Android Auto

Visit our YouTube channels:

Bloomberg Podcasts

Bloomberg Television

Bloomberg Originals