Sam Altman of OpenAI Cautions on AI Development Risks

Summary

• Altman calls for a more measured approach to AI innovation to prioritize safety.
• OpenAI is developing safety protocols before advancing training processes.
• He supports federal regulations while urging industry self-regulation.

Sam Altman, the CEO of OpenAI, has advocated for a deliberate approach to AI development, emphasizing the importance of implementing safety measures before further advancements. He believes that developers should not wait for regulatory frameworks but should proactively ensure the safety of their technologies.

In a recent tweet, Altman expressed his concerns over the rapid pace of AI evolution, highlighting the potential risks of losing control over AI systems or allowing power to concentrate within a single entity. He warned that such scenarios could lead to detrimental outcomes for society, stating, "two ways AI progress could go very badly and that we must avoid."

There are two ways AI progress could go very badly and that we must avoid.

First, we could lose control of the future to AI. This is unacceptable; we are unapologetically on Team Humanity, and AI must always serve people.

— Sam Altman (@sama) September 14, 2026

Altman elaborated on the need for AI to serve humanity, insisting that alignment and safety measures must outpace technological capabilities. He expressed concern about the risks associated with a single company or individual gaining too much power, which could undermine societal values.

He further clarified that while he does not advocate for halting progress, there is a need for a slower pace that allows for thorough safety evaluations. OpenAI is now taking steps to create explicit safety protocols ahead of significant advancements in their AI models, rather than only after models are completed.

Altman encouraged collaboration across the industry to establish shared safety standards that align AI development with human interests. He also stressed the importance of instilling public trust in AI firms as they navigate rapid advancements.

While Altman supports federal safety regulations and the role of independent auditors, he emphasized the need for companies to take responsibility for their developments without waiting for government action. He mentioned OpenAI's past consultations with Congress regarding potential coordinated efforts to slow down industry competition.

His comments come in the wake of recent cybersecurity incidents involving OpenAI and Anthropic, where both companies faced challenges that underscored the importance of robust safety measures in AI development.