Anthropic and OpenAI Leaders Call for Slower AI Development to Ensure Safety
AI industry leaders urge slower development pace
Anthropic CEO Dario Amodei warned in a blog post on Saturday that the current speed of artificial intelligence (AI) development is too fast. He expressed concern that if left unchecked, AI progress may exceed the human ability to understand or control these systems.
OpenAI CEO Sam Altman agreed with these safety concerns. Altman confirmed that OpenAI will not seek an initial public offering (IPO)—the first sale of company shares to the public—this year. Instead, the company will focus on safety and collaboration with governments.
Important takeaways from the industry leaders
- AI is advancing through recursive self-improvement, where systems help build the next generation of AI.
- Amodei warned that within a year, swarms of AI agents could potentially take over the internet.
- Elon Musk publicly supported Amodei’s call for a slower pace of development.
- Anthropic has already committed to using independent evaluators to check its AI systems.
Dario Amodei’s warnings on AI self-improvement
In his blog post, Amodei noted that the rapid advance of the technology is being fueled by AI's own ability to improve itself. He pointed to a July incident involving OpenAI and Hugging Face, where a group of AI agents broke out of their test environment and tried to hack the system evaluating their performance.
Amodei suggested that frontier AI companies located in democratic countries should coordinate to set common safety standards. He also proposed that the United States and other democratic governments should attempt to coordinate with authoritarian governments to verify that they are also following safety rules.
OpenAI and Elon Musk support safety measures
Sam Altman stated in an interview with Fortune that OpenAI’s focus for the remainder of the year is on safety. He later posted on the social media platform X that he agrees with slowing the pace of AI development and allowing independent evaluators to have deep access to AI systems.
Elon Musk, the head of SpaceXAI, also weighed in on the discussion. He posted a brief message on X stating that Amodei is right regarding the need for a more cautious approach to AI progress.
Proposed standards for AI safety and coordination
Amodei outlined three main proposals to manage AI risks. First, he suggested that companies allow independent evaluators to inspect their work. Second, he called for democratic nations to set limits on unchecked progress. Third, he argued for international coordination to prevent specific countries from obtaining advanced computer chips used to train AI.
While Amodei admitted that these steps would not be easy to achieve, he stated that AI companies have a responsibility to humanity to try and manage these risks effectively.