80 calls with 40 companies reveal the 7 biggest blockers to enterprise AI adoption
alex_verem · x · 2026-11-20
An AI consulting team reviewed every client call since July — 80 calls, 40 companies, 16 of them PE firms or PE-backed — and distilled 7 recurring adoption blockers: not enough people who can build it (19/40), burned by a consultancy/vendor before (14/40), legacy systems in the way (14/40, e.g. "10,000 stored procedures across 2,000 repos"), security/compliance blocks (12/40), employees not using it (12/40), data scattered across systems (9/40), and AI output too unreliable without human review (9/40, "only maybe 60-70% correct"). A candid snapshot of mid-market AI friction.
More from Companies & People
- Granola cofounder on courting a cofounder 'like dating' and why real-time AI notes didn't work — soleio · 2026-11-21
- Bindu Reddy: Anthropic's IPO underwhelmed on lost lead and fear mongering — bindureddy · 2026-11-21
- Peter Yang accuses NYT of hit-piece framing in Jensen Huang leadership story — petergyang · 2026-11-21
- New York Times: Can Anthropic Revive the A.I. Trade? — nordicinst · 2026-11-21
- "Claude Opus solved video editing—you just built a wrapper on it" — AIandDesign · 2026-11-21
- Agent-scale night: 21-person company hits $100M ARR with zero support teams — LukeW · 2026-11-21