Research · curated 16 Aug 2026
Stopping a cyberattack while walking your dog - defensive AI security CEO says it's not ruff to do
First reported corma.ai
Coverage timeline
Single-source research — first reported, latest, and curated coincide.
Why it matters
Corma's study quantifies how rapidly frontier LLMs are becoming capable autonomous cyber attackers while lagging far behind on defense, a widening gap defenders must plan for as agentic attack capabilities scale.
Corma, a defensive AI security startup, published research measuring an offense-defense asymmetry by running frontier foundation models (Claude Opus 4.8, GPT-5.5, Grok 4.3, and DeepSeek V4) as both attackers and defenders across realistic enterprise environments. Across 241 scored engagements the AI attackers established a persistent, reboot-surviving backdoor in 85% of runs while general-purpose AI defenders detected only 19% of implants, with models failing to find even their own backdoors 78% of the time.