OpenAI CEO stresses need for AI safeguards to keep pace with capabilities
OpenAI CEO Sam Altman says advancing AI capabilities and safeguards together is critical as the company prepares to launch its next model, Astra.
OpenAI Chief Executive Sam Altman has said that advancing artificial intelligence capabilities and safety measures together is more critical than ever, as the pace of development towards more powerful systems accelerates.
In a post on social media platform X, Altman discussed the company's internal safety work, the timing of upcoming technological leaps, and the imminent release of its next AI model. He described the balancing act the company faces, noting that protective measures remain a priority alongside performance improvements.
"Over the summer, we have been sprinting on safety priorities; it's more important than ever for capabilities and safeguards to advance together," Altman wrote. "We have more to do but have made a lot of progress. We are also going to be launching our next model soon."
Altman acknowledged an inherent tension within the organisation between deploying cutting-edge technology and ensuring responsible governance. He said the company is proud of its work on the upcoming model, named Astra, but also believes caution is warranted at this stage of development.
"We are clearly in a phase of development where we believe caution is warranted, and we are pacing our progress to ensure that we can meet the safety standards required by new capability levels," he stated.
Altman confirmed that Astra has completed its primary training phase and represents a significant step forward in both capabilities and alignment. For subsequent models, he said the company has been deliberately slowing the timetable to allow sufficient work on safety and alignment.
He also cautioned that the broader societal effects of rapidly evolving AI tools remain largely untested and poorly understood. "AI is getting extremely capable; no one fully understands the consequences of this," Altman said, adding that managing the transition to a world with abundant and powerful AI should be a top global priority.
Altman emphasised that sustainable safety depends on society and technology evolving together, and urged the public to treat current developments with extreme seriousness.