Mastodawn

Software companies: use AI to write more code than a human is capable of comprehending

Also software companies: you’re responsible for any bugs in the massive quantities of code we forced you to generate despite it being impossible for you to comprehend it.

https://arstechnica.com/ai/2026/03/after-outages-amazon-to-make-senior-engineers-sign-off-on-ai-assisted-changes/

After outages, Amazon to make senior engineers sign off on AI-assisted changes

AWS has suffered at least two incidents linked to the use of AI coding assistants.

Ars Technica

Show thread

masukomi

@samstart the problem is AIs frequently write BS tests (just returning a pass without testing anything) or testing that the code it wrote works, not that it’s right, or never writes them to begin with. Sometimes the tests are ok, but again it’s too much for a human to review and AI review tools are just adding to the piles of bs.