Archive · 5 Oct 2026
Altman Says AI Benefits Justify Accepting Some Risk
Sam Altman leads OpenAI, a company that makes artificial intelligence.
He said OpenAI and Anthropic do not agree about how strongly AI should be regulated.
Altman believes people should be able to use AI broadly, even though it may sometimes cause harm.
He thinks the good AI can do will be much greater than those harms.
He does not want rules that promise perfect safety by greatly limiting access.
But he says some dangers, like losing control of AI, are too serious to accept.
He also does not think one lab should get to control AI and decide who benefits.
The article says OpenAI has recently backed tougher rules after incidents involving its models.
Sam Altman said OpenAI and Anthropic differ substantially on how AI should be regulated.
Altman argued society should accept some harmful outcomes in exchange for AI’s benefits and broad public access.
He rejected safety guarantees that would require restricting AI, saying he expects its benefits to outweigh harms by orders of magnitude.
Altman said he opposes extreme outcomes such as a serious loss of control over AI systems.
He criticized the idea that one San Francisco lab should control AI, while the report says OpenAI has recently supported tougher rules.
- Who
- OpenAI CEO Sam Altman, discussing differences with Anthropic and its CEO, Dario Amodei.
- What
- Altman argued that AI’s benefits and broad access justify accepting some risks, while rejecting extreme risks and control by a single lab.
- When
- In an interview with Politico’s Decoded newsletter; no date is specified.
- Where
- The interview was published through Politico’s Decoded newsletter; Altman also referred to a San Francisco lab.
- Why
- Altman said the benefits of AI and people’s ability to choose how to use it outweigh some harms, and opposed concentrating control in one lab.
OpenAI’s lighter-touch approach
Anthropic’s safety-focused approach
How much risk to accept
OpenAI’s lighter-touch approach
Altman says society should tolerate some negative outcomes because AI’s benefits and public access are worth the risk; he rejects extreme risks.
Anthropic’s safety-focused approach
Anthropic is described as the industry’s most vocal supporter of regulation, reflecting a stronger emphasis on safety-focused rules.
Who should control AI
OpenAI’s lighter-touch approach
Altman opposes giving one San Francisco lab control over AI and how its benefits are distributed.
Anthropic’s safety-focused approach
The report describes criticism of safety-focused rules as potentially entrenching large incumbents, but does not provide a direct response from Anthropic on this specific claim.
- Speaker
- Sam Altman, OpenAI CEO
- Interview
- Politico’s Decoded newsletter
- Main disagreement
- OpenAI and Anthropic differ on AI regulation
- Altman’s position
- Accept some harms for AI’s benefits and broad access
- Risks he rejects
- Extreme risks, including serious loss of control over AI systems
- Control
- Altman opposed one lab controlling AI and distributing its benefits
- Recent policy shift
- The report says OpenAI has recently backed tougher rules amid high-profile hacking incidents and rogue model behavior