News for hackers. Stuff that matters.

A comment on PoisonGPT: We hid a lobotomized LLM on Hugging Face to spread fake news. Parent comment

Re:PoisonGPT: We hid a lobotomized LLM on Hugging Face to sp... by TuringTest on Sunday July 09, 2023 @04:34PM (#36658590)
If you already know that your model contains falsehoods, what is gained by having a chain of provenance? It can't possibly make you trust it more.
[ Parent | on HN ]
Re:PoisonGPT: We hid a lobotomized LLM on Hugging Face to sp... by pixl97 on Monday July 10, 2023 @12:24AM (#36662067)
People contain a shitload of falsehoods, including you, yet you assign varying amounts of trust to those individuals.

A chain of providence isn't much different then that person having a diploma, a company work badge, and state issued ID. You at least know they aren't some random off the street.

[ Parent | on HN ]
Re:PoisonGPT: We hid a lobotomized LLM on Hugging Face to sp... by TuringTest on Tuesday July 11, 2023 @09:34AM (#36680731)
That only provides value if you have knowledge of how the diploma-issuing organisation is working.

If not, it's just a diploma from some random organisation off the street.

So, wake me up when you know how OpenAI and Google are cooking their models.

[ Parent | on HN ]