Glossary
Decision model
A model that scores a fixed set of options and returns probabilities instead of generating text, used for routing, tool selection, policy checks and grading.
Google's Gemini Decision 1 and OpenAI's Decisions API both run in a single forward pass with no decode loop, so they answer in 150–250 ms and are billed on input tokens only. They cannot explain themselves or notice that none of the options fit, so generative fallback is still needed for open-ended cases.
Related terms
AgentAgent harnessLatencyNon-human identitySandbox escape
Get the brief in your inbox
One email every morning: the stories that mattered in AI, each with a one-line "why it matters" for your role. Free, no ads, unsubscribe any time.
Published daily by AI Briefing. We never share your address.