AI Is Writing More Code—And Creating a Review Bottleneck
Founder & CEO of Speak-Y — voice-to-text for macOS, Windows & iOS. Founder & CEO of Speak-Y — voice-to-text for macOS, Windows & iOS. Founder & CEO of Speak-Y — voice-to-text for macOS, Windows & iOS.
Founder & CEO of Speak-Y — voice-to-text for macOS, Windows & iOS. Founder & CEO of Speak-Y — voice-to-text for macOS, Windows & iOS.
Founder & CEO of Speak-Y — voice-to-text for macOS, Windows & iOS. Everyone measuring AI-assisted development is reporting the same shape of result: output is up, and the cost of checking that output is up much more.
What Happened
I want to walk through the numbers, because the gap between the two is larger than most teams have adjusted for, and then say what it looks like from inside a small team where I do most of the reviewing myself. Not.
Faros AI published an engineering report built on two years of telemetry from 22,000 developers across more than 4,000 teams.
A 67.4% increase in daily pull request contexts means the interruptions do not just multiply, they arrive on someone else's schedule.
Sources: Faros AI, "The AI Engineering Report 2026: The Acceleration Whiplash" (telemetry from 22,000 developers across 4,000+ teams).
Key Details
So teams write more, and the queue in front of the reviewer grows faster than the queue behind the author. Some of that queue gets resolved by not reviewing at all, and the incident numbers follow.
Sonar, "State of Code Developer Survey 2026" (1,100+ developers, published January 8, 2026).
Founder & CEO of Speak-Y — voice-to-text for macOS, Windows & iOS.
Where does your review time actually go, and what have you changed to make it survivable?
Why It Matters
Read the first two together and you get the shape of the problem. Almost nobody trusts the output, and barely half of them verify it.
If your planning captured the first number and not the second, your team is absorbing the difference somewhere: in senior engineers' evenings, in merges nobody looked at, or in the incident count.
Throughput up a third, review time up more than four times over.
What Reports Say
Coverage of the story so far points to:
Continued reporting by HackerNoon as more details emerge
