OpenAI’s Jev Clone Could Help the Frontier Lab Stop Its Swarming Agents
OpenAI’s new Decisions API, announced at DevDay and built on GPT-6 Luna, picks fast answers from a fixed list, much like TypeSafe AI’s Jev. We explain what OpenAI announced, how it compares with Jev, what a decision model is, why it could cheaply supervise misbehaving AI agents, the numbers from the Jev Sentinel demo and what developers should test.
