Right. And when we automate work by formalizing it into verifiable, testable rules, it's called… programming. We have been doing that for decades.
Ironically I feel like our QA team is busier than ever since most e2e user-ish tests require coordinating tools that is just beyond current LLM capabilities. We are pumping out features faster that require more QA to verify.
this is just an intermediate thing until the tooling and models catch up
Progress is not always linear, Until it actually does it we can't say anything. This assumption is only peddled by AI companies to get the investments and is not a scientific assumption.
Just a couple more weeks and a couple more trillion to Altman.
Would this be a problem if you can write E2E tests just like unit tests, like with Django+playwright?