news

May 15, 2026 Launched this research blog — notes on AI-agent architectures, reinforcement learning, and human–AI interaction risk.