Hugging Face 與 Protect AI 合作半週年成果：已掃描超過 400 萬個模型，守護開源 AI 安全
Original: 4M Models Scanned: Protect AI + Hugging Face 6 Months In
As open-source AI models have grown explosively, Hugging Face has become the central hub for developers worldwide to access and share…
Hugging Face 與 AI 安全公司 Protect AI 合作滿六個月，宣布已完成平台上 400 萬個開源模型的安全掃描。此項合作旨在解決開源模型（特別是含有 pickle 反序列化風險的舊格式）中潛在的惡意代碼與安全漏洞。透過自動化安全掃描與漏洞標記，雙方共同為全球開發者建構更安全、更值得信賴的 AI 供應鏈生態。
As open-source AI models have grown explosively, Hugging Face has become the central hub for developers worldwide to access and share models. However, open-source models also bring security risks that cannot be ignored — the Pickle serialization mechanism commonly used by traditional PyTorch formats (such as `.bin` or `.pt`) is highly susceptible to malicious code injection and remote code execution (RCE) attacks. To address this challenge, Hugging Face partnered six months ago with Protect AI, a leading company focused on AI security, to integrate Protect AI's security scanning technology into the Hugging Face Hub.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Hugging Face Blog →Summaries are AI-generated; the original article is authoritative.