API Flaws Put AI Models at Risk of Data Poisoning

Hugging Face Fixes Flaw; Meta, Other Tech Giants Revoke Vulnerable Tokens
Security researchers could access and modify an artificial intelligence code generation model developed by Facebook after scanning for API access tokens on AI developer platform Hugging Face and code repository GitHub. Tampering with training data is among the top threats to large language models.

access access tokens ai developer ai models api artificial artificial intelligence code code repository data data poisoning developer developer platform facebook fixes flaw flaws github hugging face intelligence meta platform poisoning repository researchers risk scanning tampering tech threats tokens top threats training training data vulnerable