MIT's AI Decision Matrix: allocating human vs AI decisions by ambiguity and risk
mcraddock · x · 2026-10-19
MIT Sloan's Center for Information Systems Research (CISR) has published an "AI Decision Matrix" to help enterprises decide which tasks AI can be trusted with and when humans must stay involved — arguing that designing decision rights is a defining challenge of the AI era.
The framework maps decisions along two dimensions:
- Ambiguity: whether the decision is well understood, or learning matters more than precision
- Risk: how broad the consequences of being wrong are
This yields four decision types:
- Exploratory (high ambiguity / low risk): enable experimentation to learn fast, largely AI-executed
- Strategic (high ambiguity / high risk): organization- or ecosystem-wide consequences, lead with continuous human oversight
- Routine (low ambiguity / low risk): automate and monitor
- Consequential (low ambiguity / high risk): broad impact, manage for reliability
Each decision is decomposed into Frame (define it), Act (take or recommend action) and Learn (own and improve outcomes), with each step assigned to a human, human-with-AI, or AI. Adapted from Sebastian, Weill, Haskamp & vom Brocke's 2026 paper "Designing Decision Rights for AI."
More from AGI Musings
- AI today is the slowest you'll ever use — in 5 years it'll look like dial-up, says founder — cramforce · 2026-10-19
- NYT on Consciousness: Even Babies' Minds Are Undecided, So Be Humble About LLM Claims — juliusadml · 2026-10-19
- Ofir Press: AI cracked a Millennium Problem before a Billboard hit — OfirPress · 2026-10-19
- From BabyAGI to ChatGPT dots: 12 posts that map 4 years of personal agents — heyneighbor · 2026-10-19
- Prompt engineering to agentic: will this year's AI skills be irrelevant next year? — MudlarkJack · 2026-10-19
- FT essay makes the moral case for human music as Suno CEO comments enrage musicians — JMateosGarcia · 2026-10-19