Sam Altman Warns AI Could Escape Human Control

OpenAI’s Sam Altman Warns Humans Could Lose Control of AI

OpenAI CEO Sam Altman is urging the AI industry to “pace” development, warning that advanced systems could create outcomes that “go very badly” if safety and governance fail to keep up with capabilities.

In a post on X, Altman outlined two core risks he believes must be avoided as models become more powerful. The first is a loss of human control over AI-driven progress. “First, we could lose control of the future to AI,” he wrote, adding that this scenario is “unacceptable” and that the company is “on Team Humanity,” with AI required to “always serve people.”

Altman said preventing that outcome depends on ensuring that alignment and safety techniques stay ahead of progress in model capabilities. He framed this as a baseline requirement for keeping people in charge of how AI is deployed and how it shapes society.

The second risk, he said, is excessive concentration of power. Altman warned that if an extraordinarily powerful AI were controlled by a single person or company, it could be used to impose a worldview on others, producing “extremely dystopian” results.

Alongside the public messaging, Bloomberg reported that Altman told employees at a company-wide meeting that he is open to slowing the development of AI agents—potentially in coordination with other labs—though he acknowledged not all competitors may agree to pace progress.

These comments arrive amid a broader debate over advanced AI risks. A new wave of warnings has also emerged from industry insiders, including Jacob Coxon, an AI researcher who said he left roles connected to Anthropic and OpenAI and argued that leading labs are “gambling with our lives,” highlighting fears among some researchers about worst-case outcomes if safety work lags deployment.

Altman’s stance also emphasizes actions the industry can take without waiting for lawmakers. He has pointed to safeguards during training and the development of shared industry standards as tools developers can implement proactively while regulation catches up.

For crypto and digital-asset communities, the stakes are closely linked to how powerful AI is governed. AI systems are increasingly used across finance and online platforms, and Altman’s warning underscores two pressures that resonate in crypto debates: how to keep humans accountable for automated decision-making, and how to prevent critical technology from becoming overly centralized in the hands of a few entities.