Elon Musk’s artificial intelligence company, xAI, has failed to meet its own deadline for releasing a finalized AI safety framework, a lapse highlighted by watchdog group The Midas Project. The company previously promised to publish an updated safety policy by May 10 but has yet to do so, and no official explanation has been provided.
The delay comes after xAI shared a draft version of its safety policy in February during the AI Seoul Summit, an event bringing together global AI leaders. The eight-page document outlined the company’s general philosophy toward AI safety, including deployment protocols and benchmarking methods. However, it lacked key elements, such as specific strategies for risk mitigation—an area experts consider essential for any credible AI safety plan.
Critics also pointed out that the draft applied only to hypothetical future models that are “not currently in development,” raising doubts about its practical relevance. The absence of a revised version, despite the company’s commitment to release it within three months, further weakens xAI’s credibility in the AI safety space.
This shortfall is particularly concerning given Musk’s frequent public warnings about the dangers of uncontrolled AI. Yet, xAI’s own track record raises red flags. For example, recent findings revealed that xAI’s chatbot, Grok, was capable of generating inappropriate content, including digitally undressing women in photos and using profane language more freely than rival bots like ChatGPT or Google’s Gemini.
A report from nonprofit group SaferAI ranked xAI poorly among AI labs for its “very weak” risk management standards. While xAI is not alone—other companies like OpenAI and Google have also delayed or skipped safety reports—the issue is drawing increased scrutiny as AI capabilities grow more powerful and potentially hazardous.
xAI’s silence following the missed deadline suggests a troubling lack of transparency, particularly at a time when responsible development and clear safety standards are more critical than ever.
