250 bad docs can poison an LLM
Anthropic just revealed that poisoning an AI doesn't require millions of malicious files—just around 250. Across models ranging from 600M to 13B parameters, attack success relied on a fixed document count rather than dataset percentage, easily planting backdoors that trigger gibberish outputs. Why it matters: Hijacking frontier models during pretraining is vastly cheaper and easier than the industry previously assumed.