April 14, 2026
Vibes vs Metrics: Fight!
90% of CEOs Say AI Changed Nothing. The Other 10% Have a PR Team
Commenters split: costly hype or rare lightning strikes
TLDR: A big survey says 9 in 10 executives saw no AI productivity boost, while many companies still trumpet gains without measuring them. The comments split between “costly scam” and “bursty but real,” uniting on one demand: less hype, more proof—because budgets, jobs, and sanity are on the line
Forget the glossy AI press tours—commenters are dragging them back to earth. The headline stat—an NBER survey saying 9 out of 10 execs saw zero impact—landed like a dunk tank, soaking every “up-and-right” slide. Readers paired it with the author’s receipts: earnings calls with no numbers, McKinsey claims without measurement (only 20% track!), and a LeadDev finding that 60% of engineering leaders saw no meaningful boost. The crowd’s verdict: fewer feelings, more findings. Bonus outrage for the hidden costs listed—padded promises, manager burnout, abandoned pilots—aka “PR-driven productivity.”
Then the knives came out. Skeptics like austin-cheney said AI is a “people problem” that shifts work downstream and nets out to zero—“big bills, no gains.” josefritzishere went nuclear: studies show an 18% productivity drop, “call it a scam.” zingababba added brain-melt vibes and pointed to layoffs. On the other side, marstall swore the boost is real but bursty—20% faster for 20% of the day, or “1000%” on two unicorn days a month. clay_the_ripper played history teacher: early tech always looks messy; innovators win first. The meme of the day? “Metrics vs vibes,” with a side of “PR-as-a-Service” and AI interns who never got onboarded
Key Points
- •An NBER survey of nearly 6,000 senior executives found nine out of ten reported zero AI impact on employment or productivity over the prior three years.
- •McKinsey’s 2025 research says 62% of organizations report 25%+ AI-driven productivity gains, yet only 20% of engineering teams measure AI’s impact with metrics.
- •The LeadDev 2025 report of 600+ engineering leaders found 60% said AI hadn’t meaningfully boosted team productivity, with most seeing only small gains.
- •The article describes how productivity claims often inflate as they move up management layers, lacking concrete measurement and specifics.
- •Prematurely treating AI as productive can raise delivery commitments, cause manager burnout, reduce adoption due to unplanned learning curves, and undermine future implementation.