tay1or

You can't tell if a human has reviewed AI-assisted / generated code. But you can tell if an AI wrote a comment or commit message. The evaluation section of the paper goes over in-depth the findings: https://github.com/ActuallyTaylor/strata/blob/main/paper/Tay...

you got me, this is the real point of the list

The data was collected for a paper that was attempting to find a lexical difference between human contributions and LLM contributions. The reason for this was because repository maintainers need to make sure submitted code has been at least reviewed by a human. By detecting unattributed AI submissions, it can tell maintainers that they need to pay extra attention to the repositories.

This surprised me as well when I wrote the paper! I started data collection for the AI assisted repositories on May 1, 2025 which is the release data of claude code. The reasoning behind this is noted in the paper but can be broken down to, it would be easier to find a large quantity of AI repositories after vibe coding took off. The majority of these repos being AI adjacent topics was an interesting find, but supports the idea that a large chunk of our LLM advancements are being helped by LLMs.