Post
HackMag — Top-notch cybersecurity magazine
hackmag.com
did:plc:o6zj77adynrtx7d5kv2t5imm
🟢 Anthropic: As few as 250 malicious documents in the training data can poison an LLM
🗨️ Anthropic specialists, together with the UK government’s AI Safety Institute, the Alan Turing Institute, and other acade…
#news
https://hackmag.com/news/poisoning-llm?utm_source=bluesky&utm_medium=social&utm_campaign=repost_hackmag_to_socials
2026-03-23T22:40:04.651Z