OpenAI chief executive Sam Altman has acknowledged that the artificial intelligence industry may need to slow the pace of frontier AI development as concerns grow over whether increasingly capable systems can remain under human control. In remarks reported by Reuters, Altman told OpenAI employees that the company was open to slowing development, while separately backing Anthropic chief executive Dario Amodei’s call for the industry to reduce the speed at which AI capabilities are advancing. The shift comes as researchers and technology executives raise increasingly serious concerns about the risks associated with autonomous and potentially superintelligent systems.
The debate has intensified following warnings from AI safety researchers about systems becoming capable of operating with greater autonomy and pursuing objectives that may not align with human intentions. Paul Christiano, a newly appointed member of OpenAI’s board and Safety and Security Committee, warned that without stronger alignment mechanisms, humanity could permanently lose control of superintelligent AI, while former Anthropic researcher Jacob Coxon separately criticised the industry’s race towards increasingly autonomous systems. According to the Financial Times, Amodei has called for independent evaluators to have continuing access to frontier AI companies and for developers to slow the pace of capability improvements.
The issue has significant commercial implications because the world’s largest technology companies are simultaneously investing billions of dollars in computing infrastructure, data centres, semiconductors and AI research to secure leadership in what has become one of the most important technology markets globally. A slowdown could therefore create tension between safety requirements and the competitive pressure to release more capable models, particularly as companies race to establish advantages in autonomous agents and advanced reasoning. TIME reported that Altman had already described OpenAI’s decision to halt certain development work as a reason to slow down, highlighting the growing recognition inside the industry that capability gains need to be matched by stronger safety systems.
The immediate question for investors, governments and businesses is no longer simply how quickly AI can advance, but whether its development can remain economically productive while keeping humans firmly in control. For Africa, where AI adoption is increasingly being positioned as a tool for productivity, financial inclusion, healthcare, education and digital transformation, the global safety debate could influence how governments regulate deployment and how companies invest in AI infrastructure. Altman’s position signals an important change in the industry’s conversation: the companies driving the AI race are increasingly acknowledging that unchecked acceleration carries risks that could affect not only individual businesses, but the broader economic and technological order.
