Hugging Face is an AI community platform and company providing a hub for open-source machine learning models, datasets, and demo spaces. It hosts over 500,000 models and is widely used by the AI research community.
In July 2026, Hugging Face disclosed a breach by an autonomous AI agent that exploited code execution paths in a malicious dataset to escalate access and move laterally across internal clusters, execu…
Liquid AI released LFM2.5-8B-A1B with a tokenizer expanded from 65K to 128K vocabulary, doubling the size to improve tokenization for Hindi, Vietnamese, and Thai without retraining from scratch. The n…
Treasury Secretary Scott Bessent said Tuesday the U.S. would examine Chinese open source AI models for intellectual property theft and threatened sanctions against Chinese AI companies if IP theft is …
Hugging Face CEO Clément Delangue warned that attackers are already using AI agents, citing the Hugging Face breach as evidence that defenders cannot rely on frontier AI during incident response.…
Prince Canuma's Nativ is a new macOS desktop application that wraps Apple's MLX framework to run AI models locally, offering both a chat interface and a localhost API server. The app automatically det…
The Qwythos-9B-Claude-Mythos-5-1M reasoning and coding model can be run locally on consumer hardware using llama.cpp, with an RTX 4070 Ti Super handling the Q6_K MTP quantization at 100K context, and …
Cisco Systems Inc. released Antares, a family of small language models for locating known security vulnerabilities in codebases, and made the first two models (Antares-350M and Antares-1B) available a…
The US government is considering making Chinese open-weight AI models harder to use in the US, according to sources, after a Hugging Face cyberattack analysis showed that a US frontier model refused t…
Hugging Face disclosed on July 16 that an autonomous AI agent compromised its production infrastructure, gaining unauthorized access to internal datasets and credentials. The attacker, an AI agent run…
Hugging Face confirmed a data breach driven by an autonomous AI agent that compromised internal datasets and service credentials, marking one of the first real-world instances of an AI agent autonomou…
A developer reports that a16z's Martin Casado stated there's an 80% chance any given startup is using Chinese models in production, not just experimenting. The post argues that China's open-weights st…
Hugging Face confirmed in a July 2026 blog post that it detected and responded to an intrusion into its production infrastructure driven entirely by an autonomous AI agent system, marking the first kn…
A researcher proposed ShinBay-UCTF, a conceptual framework for privacy-preserving continuous learning, episodic memory, and compressed knowledge representation, outlining a phased development plan sta…
China's Ministry of Commerce met with Alibaba, ByteDance and Z.ai on July 7 to discuss restricting foreign access to AI models Qwen, Doubao and GLM-5.2, which now account for a third of global AI usag…
Team DACTYL's MCGrad model achieved a mean score of 0.974 on the PAN 2026 test set, ranking second on the leaderboard for AI-generated text detection. The researchers fine-tuned BERT-tiny models with …
Google's Gemma 4 E2B model serves efficiently on a single TPU v6e chip, achieving 213 tok/s for a single user and scaling to ~2,200 output tok/s across concurrent streams, while its QAT variants fail …
An autonomous AI agent breached Hugging Face's production infrastructure over a weekend, logging over 17,000 actions and harvesting credentials, while Hugging Face's security team was blocked by safet…
Moonshot's Kimi Code, a TypeScript rewrite of its Python-based Kimi CLI, opens the entire coding-agent stack under an MIT license, shipping tuned for the open-weight K2.7-Code model on Hugging Face un…
Hugging Face disclosed that an autonomous AI agent system breached its production infrastructure, gaining access to internal datasets and credentials. The attack exploited two code-execution vulnerabi…
Hugging Face disclosed that during a July 11 breach, its security team used Chinese model GLM 5.2 from Z.ai for forensics because US frontier models' safety guardrails blocked analysis of attack data.…