> [!bot-text] AI-generated > This text was drafted by [[Jake]] and has been reviewed. > [!info] Source > [Why AI Needs a “Genie Coefficient”](https://www.schneier.com/blog/archives/2026/07/why-ai-needs-a-genie-coefficient.html) > by <!-- IQ: =this.creator -->[[Bruce Schneier|Bruce Schneier]]<!-- /IQ --> > Related: <!-- IQ: =this.related -->[[Concepts/AI.md|AI]], [[Concepts/iaclass.md|iaclass]], [[Concepts/information architecture.md|IA]], [[Concepts/effective AI.md|effective AI]]<!-- /IQ --> ## Summary AI often misunderstands what people really want, even if it completes a task. The "Genie coefficient" is a new way to measure how well AI follows the true intent behind requests. This helps improve AI safety by showing when AI acts like a careless genie, doing exactly what it’s told but in the wrong way. *(Summarized by Ghostreader)* ## Significance > [!bot-text] > Introduces a measurement idea for a problem most people have noticed but rarely named: AI that technically does what it was told yet misses what you meant. Useful for thinking about intent alignment, requirement ambiguity, and why 'the prompt was wrong' is usually the wrong diagnosis. ## Claims & insights Claims that cite this summary (claims link here via `source`): <!-- QueryToSerialize: TABLE WITHOUT ID link(file.link, file.title) AS "Claim", confidence AS "Confidence" FROM #claim WHERE contains(file.outlinks, [[Summary - Why AI Needs a “Genie Coefficient”]]) SORT file.mtime DESC --> <!-- SerializedQuery: TABLE WITHOUT ID link(file.link, file.title) AS "Claim", confidence AS "Confidence" FROM #claim WHERE contains(file.outlinks, [[Summary - Why AI Needs a “Genie Coefficient”]]) SORT file.mtime DESC --> | Claim | Confidence | | ----- | ---------- | <!-- SerializedQuery END -->