Hacker News Reader

They do scale if the examiner is an LLM.

The article mentions study content that AI has got confidently wrong. That’s going to be some crap markings.

If the ai is good enough to do the homework, it should be good enough to check the homework.

I mean some people are good at Tetris and some aren’t. Tetris can tell you which ones are which.

That’s the problem though - the article has an example of the AI screwing up the homework and the students don’t notice even when warned.