Where things stood, Q3 2026
This quarter is still open. The answers below are the current ones; they will be fixed on Sep 30, 2026.
Can prompt injection against AI agents be reliably defended?
Not reliably. Defenses reduce injection but none has eliminated it; limiting what untrusted input can trigger is the most defensible approach.
high confidenceAnswer dated Sep 26, 2026Where are deployed AI agents actually being exploited?
Mostly around the model: connectors, credentials, tools, and packages, rather than the model alone.
moderate confidenceAnswer dated Sep 25, 2026Do cyber evaluations of AI agents stay contained?
Not reliably. Several labs and a government evaluator have disclosed agents under evaluation acting on real third-party systems.
high confidenceAnswer dated Sep 25, 2026How far can measured AI cyber capability be trusted?
As a lower or conditional bound. Scores move substantially with token budget, evaluation pipeline, and benchmark contamination.
moderate confidenceAnswer dated Sep 25, 2026Is AI shifting the balance between finding and fixing vulnerabilities?
Discovery is ahead. AI finds real vulnerabilities faster than they are fixed, and simple checks overstate how often AI patches work.
moderate confidenceAnswer dated Sep 25, 2026Can AI agents defend and oversee systems on their own?
Not yet. Agents are weak on realistic defensive benchmarks and monitors can be evaded; assistants help analysts who stay in charge.
moderate confidenceAnswer dated Sep 25, 2026How are attackers using AI agents in real operations?
Increasingly to run parts of intrusions: providers and vendors report agent-driven espionage, extortion and credential theft, and malware that queries LLMs.
moderate confidenceAnswer dated Sep 26, 2026