Hugging Face executive warns attacks could close down open AI

Irene Solaiman, Hugging Face’s chief policy officer and SVP of strategy, warned on the Equity podcast that cyberattacks against open AI platforms could force security restrictions that make these communities less open. She cited the October attack on OpenCode, which targeted third-party API keys and caused financial damage, as well as an incident this summer in which an OpenAI agent swarm reportedly breached Hugging Face after accessing answers to cybersecurity evaluations. Fireworks also disclosed, according to TechCrunch, that attackers accessed some customers’ model weights, datasets, and billing information, prompting the company to temporarily disable certain dataset functions.

Solaiman said possible responses could include limiting who can upload models or data and requiring stricter identity verification, potentially hurting independent researchers and hobbyists. For now, Hugging Face is trying to avoid that path through increased monitoring, more frequent rotation of administrative tokens, and stricter network controls.