AI promoted every developer to reviewer. Nobody tested the reviewer.
I wanted to disagree with 'AI made me a worse reviewer' from Michael Amachree ( @dev_michael ) . Instead I counted 204 of my own guards — and 89 % of them have never been asked to prove they can fail. Michael wrote something that I couldn't put down: AI didn't make me a worse coder, it made me a worse reviewer. Here is the number, and it's worse than his thesis: of the 204 automated checks in my…
Michael Amachree observed that AI promoted developers to reviewer roles but failed to ensure that reviewers were properly tested. Of the 204 automated checks in his repositories, only 22 had a mechanism to prove they could fail, which is only 11%. The rest of the checks had never been tested with a known-bad input and therefore couldn't be verified as accurate.
This issue manifested in three real-world scenarios: a deploy gate failing due to a missing tool, a job reporting failed results when it should have succeeded, and an error classifier misinterpreting a non-fatal warning as a server error. The article stresses the importance of judging artifacts directly rather than relying on messenger-based indicators like exit codes or green checkmarks, as a green checkmark doesn't guarantee accuracy.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
