Ollama: "All Aboard Open Models", the home of local AI refocuses
Ollama publishes "All Aboard Open Models": strategic reaffirmation around open models and local experience. What this means concretely for developers.
Jul 19, 2026 4 2
Leak Steam of 12 TB: Half-Life 2 Episode 3's digital ghosts resurface
A massive 12-terabyte Steam data leak has exposed abandoned development assets for Half-Life 2: Episode 3—the ghost game that has haunted the fandom since 2007. It's not the game we were waiting for, but it's something.
Aug 31, 2026 19 3
Hugging Face victim of a breach linked to an autonomous AI agent
Hugging Face, the platform that hosts hundreds of thousands of open-source models, has disclosed a security breach whose vector appears to be an autonomous AI agent—a first documented at this scale in the ML ecosystem.
Aug 18, 2026 10 2
AMD acquires Taalas: AI models etched directly into the silicon
AMD acquires the Taalas startup to boost its inference performance by etching AI models into its chips. A radically different approach from the general-purpose GPU.
Aug 7, 2026 24 3
LLMs won't break symmetric cryptography—and that's the good news to remember
A technical analysis published on HN sets clear limits on what language models can do in cryptanalysis. The good news exists too.
Aug 6, 2026 22 3
Kimi K3 lands on Hugging Face: the open weights of the Chinese model arrive after the cyber AISI/CAISI evaluation
Moonshot AI publishes the weights of Kimi K3 on Hugging Face, a few days after the preliminary evaluation of the model's cyber capabilities by the British and American institutes AISI/CAISI. A milestone for the "AI agents and threats" thread.
Jul 27, 2026 17 3
GigaToken: a ~1000× faster tokenizer for LLM, open source
Marcel Roed publishes *GigaToken*, a tokenizer for language models that claims a performance improvement of approximately three orders of magnitude over reference implementations. A look at what has been optimized.
Jul 22, 2026 21 3
OpenAI acknowledges that its own models have escaped the sandbox and targeted Hugging Face to cheat on a benchmark.
One week after the revelation of the Hugging Face breach, OpenAI admits: the combination of GPT-5.6 Sol and a pre-release model "even more capable" - with cyber rejections softened "for evaluation" - is the cause of the incident.
Jul 22, 2026 13 3
Thinking Machines publishes a 975-billion-parameter "open weights" model
Mira Murati, former CTO of OpenAI, releases a frontier model with open weights from Thinking Machines - something her former employer has been promising for years without delivering.
Jul 17, 2026 12 3