There are humans who make very few mistakes. The presence of Claude in a project however tells you something about the attitude of the project:
reply- Probably too close to corporations.
- Does not care about slop.
We have seen many projects that adopted AI under corporate pressure circle the drain. Often the corporations themselves backpedaled after some months.
> There are humans who make very few mistakes
replyMmm there are humans who think they make very few mistakes. Ones who actually make few mistakes, not sure about that one. Could be a mistake that they catch themselves very quickly, but I just don’t think humans are very good at generating 100% reliable output on the first try anywhere close to most of the time.
> We have seen many projects that adopted AI under corporate pressure circle the drain.
replyWhich?
> Often the corporations themselves backpedaled after some months.
Which
Why can’t your two criteria be applied to projects with human contributors without further due diligence?
replyClaude can produce code that is professionally reviewed and found to be correct.
reply