LLM Poisoning Breakthrough
A joint study by Anthropic, the UK AI Security Institute and the Alan Turing Institute shows you only need about 250 malicious docs to slip a “backdoor” into any large language model—whether it’s 600 M parameters or 13 B, size and data volume don’t save you.
Diwali Deal Alert
If you’re itching to level up, there’s a 20% festive discount on all live courses with code AI20—hit up Krish Naik’s team for enrollment or questions!
Watch on YouTube
Top comments (0)