The widely-cited 95-percent generative-AI-pilot-failure framing (MIT Sloan Management Review and Boston Consulting Group adoption-research streams, 2025-2026) is methodologically defensible for the enterprise cohort the research sampled (large firms with dedicated AI functions, 12-to-18-month evaluation windows, scaled-production-deployment success definition) and materially misrepresents small-firm pilot dynamics. The 1-to-50-person operator cohort has a different failure-mode catalogue (tool-assigned-to-wrong-person, rewrite-cost-exceeds-savings, client-rework-from-AI-deliverable, line-item-stack-compounded-and-cancelled, sporadic-use-no-routine) and a different success definition (90-day payback at actual hourly rate; deliverable quality reaching the client without disproportionate rework; routine fit documented for handover). A three-question Monday-morning small-firm pilot test (payback, deliverable quality, routine fit) checked at 30 days and 60 days is the operator's actual evaluation instrument and replaces the enterprise 12-to-18-month evaluation cycle that the 95-percent number is measured against.

Operators register pillar piece on pilot-evaluation framework for 1-to-50-person firms. 45-day cadence calibrated so the first review falls within the typical 30-to-60-day pilot decision window. Trigger conditions for status changes: (1) MIT Sloan, BCG, or a comparable adoption-research stream publishing small-firm-specific (1-to-50-person) pilot-failure data inside the review window (would either confirm the structural argument that the enterprise framing misclassifies, or refine the small-firm failure-mode catalogue with new evidence — keep Holding either way unless the evidence directly contradicts the load-bearing claim); (2) a small-firm operator survey from Stripe Atlas, Brex, Ramp, or equivalent SMB-spend-and-adoption data publishers producing pilot-outcome data at the cohort level that materially diverges from the five failure modes listed (would refine the catalogue and could move toward Partial if a sixth or seventh mode dominates the new data); (3) a major foundation-model provider publishing operator-cohort case studies with attributable revenue impact at the 1-to-50-person scale (would harden the success-definition argument by establishing what small-firm pilot success looks like in the public record); (4) a viral re-citation of the 95-percent number with new methodology that does include small firms (would update the source-document discussion in the piece). Sibling claim: AM-146 (three questions for CIOs about agentic AI accuracy claims) addresses the analogous misread problem in the enterprise cohort.