Read Sam Altman's plan for OpenAI as it enters its 'third phase'

OpenAI has entered its 'third phase,' according to CEO Sam Altman and Chief Scientist Jakub Pachocki, shifting focus from research and product rollout to making advanced AI accessible, safe, and beneficial worldwide. In a blog post, they outlined three main goals: building an automated AI researcher, accelerating the economy, and providing a personal artificial general intelligence (AGI) to everyone. The company stresses that AI should assist humans without replacing them, warning against unchecked automation that could pose risks. OpenAI’s plan highlights the need for global collaboration to manage AI risks, including potential slowdowns in frontier model development if necessary. Altman and Pachocki argue for a future where AI capabilities are widely distributed rather than concentrated in a few institutions. They also emphasized alignment with human intent and control as critical to safe AI deployment. On the same day, OpenAI confirmed it had confidentially filed for an initial public offering (IPO), though the company noted it may take time before shares become publicly available. The announcement aligns with broader industry discussions, as Anthropic researchers recently called for pausing AI development to allow societal structures and alignment research to catch up. The blog post reflects OpenAI’s ambition to democratize AI while addressing safety concerns, positioning the company as a leader in shaping the technology’s future. The third phase marks a transition from foundational research and product launches to broader societal integration and governance.