OpenAI launches autonomous AI agents amid model safety pause

OpenAI CEO Sam Altman unveiled the new AI agent technology at the company's developer conference on Tuesday, positioning it as a competitor to Meta's recently launched personal agent Muse. The Dot agents are designed to work continuously on behalf of users, handling ongoing tasks without requiring constant direction or input.

"These agents will function like an AI helper that always has your back, drawing inspiration from the intelligent assistants we've seen in popular films," Altman said at the event. The announcement represents a significant push by OpenAI into autonomous AI systems that can operate independently within defined parameters.

The product launch comes just one day after OpenAI announced it would delay the rollout of a more advanced model following concerns raised by the company's own research team. The decision reflects mounting pressure across the technology industry to slow the pace of AI development and allow safety protocols to keep pace with capability advances.

The move signals OpenAI's attempt to balance innovation with caution as the AI sector grapples with broader questions about deployment speed and oversight. Industry observers note that safety considerations have become increasingly central to how major AI developers approach new releases.