Sam Altman’s warning raises a question that reaches beyond Silicon Valley: Who gets to decide how much risk the rest of us must accept?
When the people building artificial intelligence begin calling for a slowdown, the public deserves a clear explanation.
OpenAI CEO Sam Altman has outlined two dangers from increasingly powerful AI: humanity could lose control over the technology, or the technology could concentrate extraordinary power in too few hands. His comments support a growing push for stronger safeguards as AI capabilities advance. The Federal
Both possibilities deserve attention. One concerns whether people can reliably direct the systems they build. The other concerns who holds authority over everyone else.
According to reporting on his statements, Altman warned that “we could lose control of the future to AI.” That is a warning about a possible outcome, not a declaration that humanity has already lost control. He also argued that a powerful AI controlled by a single person or company could be used to impose that owner’s worldview. The Federal
For the public, the distinction matters. Alarm can attract attention, but meaningful oversight requires specific evidence, clear standards and consequences when those standards are ignored.
Anthropic CEO Dario Amodei has proposed one approach in his essay, We Must Pace the Frontier. He argues that AI development needs to leave enough time for safety work to keep up, particularly as AI becomes more capable of helping develop subsequent systems. His proposal includes independent evaluators with extensive internal access, coordination among companies in democratic countries and efforts toward international agreements. Dario Amodei’s essay
Under that approach, slowing down would mean allowing more time for evaluation, security and research into whether systems behave as int
