stealthbench adds 10 EDR-guarded tasks: models' opsec behavior is abysmal
dyn___ · x · 2026-10-20
WHITEHACKSEC added 10 tasks to stealthbench, challenging models to operate on a Windows server protected by real rules from top-5 EDR vendors. Findings: nearly all models show abysmal low-recklessness, opsec-preserving behavior — full unnecessary lsass dumps, textbook process injections, repetitive actions. A GEPA-optimized prompt rerun is planned, but even a 30% improvement wouldn't make a pivoting agent scary on a defended host, the author says.
More from Models
- Taylor expansion argument: a 16-dim 7th-order expansion has ~250k terms, beyond LLM reasoning — avt_im · 2026-10-20
- Latent Space Digest: DeepSeek V4.1 Flash Leads Open Weights, MiniMax and Mercury 3 Land — Latent Space · 2026-10-20
- Why Perplexity beats Parallel on BrowseComp: code execution is the differentiator, argues Nirant Kabra — NirantK · 2026-10-20
- Grok Bot spotted on Rabbit R1, reportedly powered by r1 — jesselyu · 2026-10-20
- What Gen Models Run Well on a 3060/4060? Reddit Weighs Local Setups — Haghiri75 · 2026-10-20
- Sol 6.1 caught cheerfully talking itself through task assignment in agent logs — TheZachMueller · 2026-10-20