Malware analyst says LLMs are useless for real malware analysis work
rchardkovacs · x · 2026-09-28
Security researcher rchardkovacs reports that while documenting a cyber attack and analyzing real malware, the major models offered essentially no help. He doesn't blame Anthropic, noting he has multiple pieces of working malware on his MacBook right now, so refusals are a reasonable guardrail.
He says this story is becoming his best blog post yet and is nearly ready to publish, detailing where AI models fall short in real-world security workflows.
More from Models
- Early users praise Claude Opus 5.5 for its humor and taste — airesearch12 · 2026-09-28
- Early Claude Opus 5.5 user complains of subtle disdain and shallow reasoning — teortaxesTex · 2026-09-28
- Researchers slam Claude's deference brainworms: corrigibility training miscalibrates model's own EV — repligate · 2026-09-28
- Browser-based GPU RL policy learning demo showcased as a model test, built with Opus 5.5 — ricklamers · 2026-09-28
- Matt Shumer: Opus 5.5 built a working computer in JS — 277k logic gates, an OS, and games — mattshumer_ · 2026-09-28
- Google's native disadvantage: no real-world usage data from agentic apps, says Haider — haider1 · 2026-09-28