Stories by Aedelon

Anthropic's Mythos leak: 3k files in a public CMS, and what the docs revealed

Testing Nvidia's FP4: Running 70B LLMs on a Single RTX 5090 with Real Benchmarks

China's AI progress by the numbers: GLM-5 benchmarks, robotaxi, and Huawei chips

AI Broke ARC-AGI-2 at 84.6% – But the Key Trick Is from 1972

Pure LLMs Score 0% on ARC-AGI-2. Why the Third Wave of AI Looks Like the First

RWKV-7 beats Llama 3.2 with 3x fewer training tokens and formally exceeds TC^0

Reasoning Models Fabricate 75% of Their Explanations (ArXiv:2505.05410)

0.1% synthetic data is enough to degrade AI models (Nature, 2024)