Matthew Skelton on Nostr: "Remember that the outcomes of Large Language Models are not designed to be true — ...

"Remember that the outcomes of Large Language Models are not designed to be true — they are merely designed to be statistically likely. " - ERYK SALVAGGIO

This should basically exclude the use of LLMs for entire classes of user-facing services.

#GenAI #LLM

https://cyberneticforests.substack.com/p/a-hallucinogenic-compendium