Hacker News

points by BrokenCogs 9 hours ago | 0 comments

Isn't this easy for frontier 'labs' to fix? If the user requests lots of writing and prose, then use an auxiliary model which has been trained on 'award-winning' writing instead of all the crap on the internet?

[–] FeteCommuniste 9 hours ago

Award-winning writing has all kinds of different styles. The average of them will probably get you some boring median soup once again, though maybe a bit less bland than the LLMs trained on oceans of LinkedIn and Reddit slop.

[–] ThrustVectoring 2 hours ago

Even if it doesn't, the effect of so much writing getting pushed out the door through default LLM settings is going to make it feel same-y and bland in short order.