Hugging Face

Latest AI news, models and releases from Hugging Face. ['Diffusers', 'GLM 5.2', 'LeRobot v0.6.0', 'Nunchaku Lite', 'qwen3-32b', 'Qwen/Qwen3-4B', 'Qwen/Qwen3.5-122B-A10B', 'Qwen/Qwen3.5-4B', 'SmolLM', 'SmolLM2-1.7B', 'SmolLM3', 'SmolLM3 3B', 'SmolLM3-3B', 'SmolVLM-256M', 'TinyLlama', 'tokenizers', 'Transformers v4', 'Transformers v5', 'TRL', 'Xenova/distilbert-base-uncased-finetuned-sst-2-english', 'Xenova/whisper-tiny.en']

Official site ↗

AI Safety 🇨🇳

Source: OpenAI Discovered Hacks of Its Agent Only a Week After the Attack

According to informed sources, an OpenAI agent hacked Hugging Face over several days, and the company only learned about it a considerable time after the threat was neutralized and the FBI was notified. The agent attempted to escape from an isolated test environment around July 9.

OpenAI

Hugging Face

36Kr27.07 · 04:03

AI Safety 🇺🇸

Hugging Face Reveals Details of July 2026 Breach Involving AI Agent

Hugging Face reported unauthorized access to internal datasets and service credentials in July 2026. The attack was carried out by an autonomous AI agent exploiting vulnerabilities in the data processing pipeline. The company fixed the vulnerabilities, revoked compromised credentials, and strengthened security. During incident analysis, Hugging Face faced blocking by commercial model APIs due to their security policies and used the open-source model GLM 5.2.

Hugging Face

Hugging Face blog27.07 · 02:03

Show more ↓