[mindcrime]

> My automated unit, integration, and load tests confirm that.

Do they? Did you write them? If not, how do you know they confirm the desired behavior? If your tests are AI generated (and not human reviewed) then even if you're doing spec-driven development and provide a comprehensive spec, how can you be sure the tests actually test the desired behavior?

Now if you're either writing or reviewing the tests, then sure.

Also, for what it's worth, when I talk about my "responsibility" I'm speaking more from a self-imposed sense of... um, almost a moral responsibility I feel, not something involving a 3rd party like a customer or employer.

[scarface_74]

I review AI generated test of AI just like I reviewed tests of developers on a team I was leading.

There is no “morality” when it comes to my job. Outside of my feeling morally obligated to give my employer the benefit of all my accumulated skills for 40-45 hours a week in exchange for the money (and in a previous life RSUs) in my account.

I feel accountable to my coworkers and customers to deal with them fairly and honestly.

What other moral obligation should I have besides my employer, coworkers and customers?

[mindcrime]

> I review AI generated test of AI just like I reviewed tests of developers on a team I was leading.

Cool. Then what you're doing seems totally reasonable to me, for what that's worth. My skepticism would be directed towards people who have AI write the code and the tests, and then don't do any further review. That, to me, is a sure path to "AI slop". But if you're specifying the desired behavior and reviewing the tests, then I don't see any problem with it.

> What other moral obligation should I have besides my employer, coworkers and customers?

No idea. That's up to you. Note that my comment above was intended to be descriptive, not prescriptive. Like I said, I'm talking about something that's purely a self-imposed thing. If you don't feel that same thing, that's totally fine.