Trending PR Campaigns
Frenos’ AI and Critical Infrastructure Security Report
Frenos published a report testing how well AI models could plan attacks on industrial systems, using nearly 17,000 tests across different models and simulated tasks. The news hook was a finding that the strongest models could often devise workable plans, while still making mistakes—raising the question of whether AI could lower the barriers to attacking critical infrastructure.
Why it worked: It led with fresh, quantified tension—16,980 tests showing frontier models can plan real OT attacks—challenging “script kiddie” assumptions with hard numbers and named experts.
The lesson: Consider running a rapid, repeatable stress-test on a risky behaviour, publish pass/fail rates and variance, and brief quotable experts on the implications.
Coverage over time
How this story spread across the media
| Publication | Headline | Date |
|---|---|---|
| theregister.com | AI has 'potential' to turn script kiddies into sophisticated infrastructure attackers | Wednesday 21st October, 2026 |
| axios.com | AI is getting better at hacking power and water systems, new research shows | Wednesday 21st October, 2026 |
| theoutpost.ai | AI Models Planning Attacks on Power Grids, Water Systems | Wednesday 21st October, 2026 |