Sam Altman says OpenAI won’t go public until its models are safe

OpenAI CEO Sam Altman unveiled an always-on AI agent called Dots during the company’s annual developer conference on Tuesday. The Dots agents are designed to proactively complete ongoing tasks on behalf of users, making them a competitor to Meta’s popular personal agent Muse. Altman described the Dots agents as "an AI helper that always has your back," inspired by the futuristic depictions of AI he witnessed in movies.

Earlier that day, OpenAI had announced it was postponing the release of a new AI model due to security concerns raised by its researchers, a move that coincided with calls within the industry to slow down the pace of AI development to allow time for safety measures to catch up. Altman did not address the scrapped model during his remarks, but he emphasized that the ultimate goal of AI should not be to turn people into cogs in a machine.

Instead, he argued that AI should empower individuals to have more control over their lives and provide them with tools to create, learn, and discover.

In addition to the Dots agents, OpenAI also announced the release of GPT-6.1 Sol, an upgraded version of its GPT-6 Sol model, and a premium speed tier called "Ultrafast." The company's announcements came as industry peers met with President Donald Trump in Washington. While Altman was not present due to the conference, other leaders from the tech industry, including Greg Brockman (OpenAI president and co-founder), Dario Amodei (Anthropic), Jeff Bezos (Amazon), and Elon Musk (Tesla and SpaceX), were in attendance.