Results for “"gibberish"”
1–2 of 2 articles
Best match · title matches first
The Register
It's trivially easy to poison LLMs into spitting out gibberish, says Anthropic
Just 250 malicious training documents can poison a 13B parameter model - that's 0.00016% of a whole dataset Poisoning AI models might be way easier than previously thought if an Anthropic study is anything to go on. …