zlacker

Turns out that AI is not good at summarising things:

https://futurism.com/ai-chatbots-summarizing-research

>>Andrew+(OP)
As usual the paper is dead on arrival. They tested with obsolete models and non-reasoning models.

Try again with any SOTA reasoning model (GPT-o3, Gemini 2.5 Pro, Grok 3).

>>Andrew+(OP)
It was good enough with summarizing this empty rant.

>>Andrew+(OP)
Ok, I disagree.

Out of curiosity, I used ChatGPT to make a summary of “FreeBSD vs Linux comparison”, and if came out as extremely fair and to the point, in my opinion.

>>Andrew+(OP)
There was this widely discussed paper about how using ai generated data for training will be the death of llms that was published about a week before all llms switched to using ai generated data for training and becoming twice as good as before