so work just rolled out a policy that half our sprint has to be ai generated now and nobody asked if that was even a good idea for the kind of stuff we ship and i spent two hours yesterday untangling a function the bot wrote that technically passed every test but did the actual business logic completely wrong and i only caught it because i happened to know that one edge case and i keep thinking about what happens the week i dont catch it and my lead just says explainability isnt really his department so is there something out there that actually reviews ai generated code for logic issues before it gets merged instead of just trusting the green checkmark because im starting to feel like the last line of defense here and that cant be the plan
Anonymous Ibex,