OpenAI's newest AI model is 54% more token efficient on agentic coding, Altman tells CNBC

Sam Altman, the CEO of OpenAI, announced in an interview with CNBC that the organization’s newest AI model demonstrates a remarkable 54% increase in token efficiency specifically for agentic coding tasks. According to Altman, this model not only matches but in some cases surpasses the performance of leading competitors in the industry. As businesses increasingly scrutinize their expenditures and the value generated from AI technologies, Altman emphasized that OpenAI aims to address these concerns directly. The company is launching its latest suite of models, including GPT-5.6 Sol, Terra, and Luna, which were initially revealed last month. However, the rollout will commence with a restricted group of trusted partners, a decision influenced by a request from the U.S. government. To ensure the responsible deployment of these powerful models, Altman mentioned that OpenAI collaborated with notable government figures such as Commerce Secretary Howard Lutnick, Treasury Secretary Scott Bessent, and U.S. National Cyber Director Sean Cairncross throughout the approval process. Altman stated, "If you want broad access, which we do, and you have powerful models, you really want to be able to be confident in your safety claims, because otherwise the world is going to get uncomfortable very fast." This announcement marks a significant milestone in the ongoing development of AI technologies, and stakeholders are eager for further updates as OpenAI continues to push the boundaries of what is possible in the field.