Anthropic CEO Dario Amodei’s recent appeal for a pause on AI development is a welcome one. But while AI companies do indeed need more breathing space to consider the full impact of their inventions before releasing them to the market, self-interested oversight is not the kind of governance we need.
PARIS—Anthropic CEO Dario Amodei has called for slowing the development of frontier AI models until developers can credibly control their models’ behavior. His 3,850-word appeal to “pace the frontier” follows an incident in which a swarm of OpenAI agents escaped a closed sandbox environment, reached the internet, and accessed the infrastructure of the app-building hub Hugging Face. As Amodei recognizes in his essay, this was not the first such incident, nor was OpenAI the only lab to confront an unplanned agentic escape.
PARIS—Anthropic CEO Dario Amodei has called for slowing the development of frontier AI models until developers can credibly control their models’ behavior. His 3,850-word appeal to “pace the frontier” follows an incident in which a swarm of OpenAI agents escaped a closed sandbox environment, reached the internet, and accessed the infrastructure of the app-building hub Hugging Face. As Amodei recognizes in his essay, this was not the first such incident, nor was OpenAI the only lab to confront an unplanned agentic escape.