Re:Many SWE-bench-Passing PRs would not be merged
by cfcfcf on Wednesday March 11, 2026 @09:43PM
(#47345218)
I agree, I think this is AI, especially based on 1 and 2. It's hard to put your finger on, and I don't know if we can know for sure. It reminds me of the writing style you see on LinkedIn i.e. seemingly optimised for engagement.
If they're not already, I wonder if LLMs will get better at disguising this (avoiding the tells, inserting mistakes etc.)
I also wonder if there comes a point where we as a culture imitate this style.
Re:Many SWE-bench-Passing PRs would not be merged
by jc-myths on Thursday March 12, 2026 @12:56AM
(#47346640)
TBH, I don't like AI-generated content much too, X and many other platforms were also flooded with those, which I tend to ignore. I guess I also fall into the rabbit hole myself with the aid of AI nowadays.
Re:Many SWE-bench-Passing PRs would not be merged
by DetroitThrow on Wednesday March 11, 2026 @09:34PM
(#47345132)
Yep, there are some other tells but at least matches LLM style strongly. Worth remembering that dang is the top emdash user in HN history and might have been flagged as an AI just for that
2. Repeated short phrases ("Tests still passed. Build still passed."). This is the new "it's not x, it's y" for me.
3. Ends on a sentence that pointlessly summarises the comment.
4. One-day old account.
5. Bio says "Building AI"
6. Criticises AI despite the bio.
7. Pangram says the comment is 100% AI.
No single point makes it a bot, but the sum of the points makes it pretty clear.