Hacker News new | past | comments | ask | show | jobs | submit
grey-area | on: Progress on provable inference for large language models
But if you’re using them they’re probably already stealing your data, fake-models are the least of your problems.