arXiv Brief — 2026-09-16

LLM agents are central today, revealing new architectural risks like capability laundering and the stochastic deputy problem, while hardware vulnerabilities expand into GPU memory architectures.

September 16, 2026 · 5 min