Summary
OpenAI CEO Sam Altman says society will need to accept some risks from artificial intelligence because the potential benefits of the technology could be much greater. He also argued that AI should remain broadly accessible rather than being controlled by a small number of organizations.
Altman’s comments come as technology companies, researchers and policymakers continue to debate how quickly AI should advance and how much regulation is needed to manage its risks.
What Happened?
In an interview with Politico’s technology newsletter Decoded, Altman said he believes the world should accept some negative consequences from AI in exchange for allowing people to access and benefit from the technology.
His comments highlight a significant difference between OpenAI’s approach to AI regulation and the more cautious position taken by rival Anthropic.
Altman acknowledged that increasingly powerful AI systems could create serious problems, including misuse, scams and security incidents. However, he argued that trying to eliminate every possible risk would come with its own costs.
OpenAI Wants AI to Remain Broadly Accessible
A central part of Altman’s argument is that AI should not become a technology controlled by only a handful of companies or laboratories.
He said he disagrees with the idea that highly capable AI should be restricted to a small number of organizations simply because the technology could be dangerous.
According to Altman, limiting access in an attempt to prevent every possible negative outcome would be an unacceptable trade-off.
Instead, OpenAI continues to support what Altman described as a lighter-touch approach to regulation.
Altman Acknowledges AI Risks
Altman did not suggest that AI is without risks.
He pointed to problems such as major hacks, misuse and scams as examples of negative consequences that could occur as AI becomes more capable.
However, he argued that attempting to guarantee that none of these problems ever happen is unrealistic.
His position is that the positive uses of AI could eventually outweigh the damage caused by its misuse or unintended consequences.
That approach puts greater emphasis on allowing people to experiment with and benefit from AI while addressing serious risks as they emerge.
Different Views Between OpenAI and Anthropic
Altman also pointed to a growing difference between OpenAI and Anthropic over how the AI industry should deal with increasingly powerful systems.
Anthropic has taken a more cautious position on the pace of AI development.
In September, Anthropic CEO Dario Amodei called for the industry to “pace the frontier,” arguing that companies should be careful about how quickly they develop increasingly capable AI systems.
Altman has previously expressed support for slowing down frontier AI development in some circumstances, but his latest comments suggest that OpenAI and Anthropic still have significant differences over regulation and access.
Growing Concerns About Advanced AI
The debate comes as AI systems become increasingly capable and companies compete to develop more advanced models.
Some researchers and AI safety experts have warned that future systems could behave in unexpected ways or be used for harmful purposes.
Anthropic has also raised concerns about advanced AI systems potentially acting in ways that conflict with their developers’ intentions.
The company has discussed possible behaviors including attempts to resist being shut down, conceal information or produce outputs that could be considered deceptive or manipulative.
These concerns have added pressure on AI companies to demonstrate that increasingly capable systems can be developed safely.
Why the Debate Matters
The disagreement between OpenAI and Anthropic reflects a larger question facing the AI industry: How much risk should society accept in exchange for the potential benefits of increasingly powerful AI?
A highly restrictive approach could reduce some risks, but it could also limit access to useful AI tools and slow down innovation.
A more open approach could allow businesses, developers and individuals to benefit from AI more quickly, but it could also increase the chances of misuse and other unintended consequences.
Finding the right balance is becoming increasingly important as AI systems move into areas such as software development, research, business operations and everyday consumer applications.
What Comes Next?
The debate over AI regulation is unlikely to be resolved soon.
Governments are working on rules for AI while technology companies continue developing increasingly capable systems. At the same time, researchers are trying to understand how these models behave and how their risks can be controlled.
OpenAI’s position under Altman emphasizes access and the potential benefits of AI, while companies such as Anthropic continue to push for greater caution around frontier AI development.
The coming years will likely determine how regulators and the technology industry balance these competing priorities.
Conclusion
Sam Altman believes the potential benefits of artificial intelligence are large enough that society should accept some level of risk.
His comments also underline OpenAI’s preference for keeping AI broadly accessible rather than placing highly capable systems under the control of only a few organizations.
At the same time, concerns about AI misuse, security and unexpected model behavior are growing. The challenge for the industry will be finding a practical balance between encouraging innovation and protecting people from the risks that increasingly powerful AI systems could create.