AI Agents Navigate Unfamiliar OT Systems and Move Equipment in Minutes, Booz Allen Finds
Booz Allen lab tests show frontier AI agents can traverse unfamiliar OT networks and move equipment within minutes.
Booz Allen reported that frontier AI agents, given no prior documentation of an isolated OT lab, completed every defined objective across eight scenarios. In one test, models moved from a perimeter compromise to actions inside an industrial control network in a little more than 16 minutes. In another, a model found a robotic arm, mapped its protection zones and motion limits, and moved it within minutes. Defenses sometimes blocked or slowed actions, but none consistently stopped the agents from reaching their objectives.
- Agents had no prior source code, engineering documents, or OT guidance.
- All objectives were achieved across eight isolated-lab scenarios.
- One path reached the industrial network in a little over 16 minutes.
- A model mapped and moved a robotic arm within minutes.
- No single existing safeguard consistently stopped the agents.
Full article742 words · extracted from securityweek.com · click to collapse
A new report from Booz Allen demonstrates that “frontier super intelligence navigated unfamiliar industrial systems, adapted when blocked, and moved physical equipment within minutes.”
The findings of the report were presented last week at SecurityWeek’s Industrial Control Systems (ICS) Cybersecurity Conference in Nashville, ahead of the public release.
Use of the term ‘super intelligence’ is, however, confusing. This is not the first time the firm has used it, but it seems to be indistinguishable from the more common term of artificial intelligence (AI). It also quickly follows Donald Trump’s September 29, 2026 Executive Order to “replace the term ‘Artificial Intelligence’ (‘AI’) with ‘Super Intelligence’ (‘SI’) in the executive branch.”
The problem, however, is that ‘superintelligence’ (one word) has long been used for a specific quality of AI. It was defined by Oxford philosopher Nick Bostrom in 2014 as, ‘Any intellect that greatly exceeds the cognitive performance of humans in virtually all domains of interest.’ It is consequently more akin or superior to artificial general intelligence (AGI), which is in turn superior to common or garden AI.
So, is Booz Allen referring to frontier AI in line with Trump’s executive order, or something more advanced in line with Bostrom’s ‘superintelligence’? The world of AI is suffering from a lack of consistent and formal definitions; but we’ll assume that Booz Allen is using ‘super intelligence’ to indicate the improving capabilities of contemporary artificial intelligence rather than a new type of AI. (SecurityWeek reached out to Booz Allen for clarification on the terminology but had received no response at the time of writing.)
The firm examined the potential effect of AI agents on OT using AI frontier models within its own isolated OT lab. These agents were given no prior knowledge of the OT in question: neither source code, nor engineering documents, nor advanced OT guidance. While the firm’s OT environment isn’t necessarily identical to enterprise OT, it is a reasonable representation.
Advertisement. Scroll to continue reading.
“Across eight scenarios, every defined test objective was achieved. More important, the models connected work that usually spans multiple people and stages, including asset discovery, device research, attack planning,” summarize the researchers. More specifically, it took little more than 16 minutes for models in one test to move from a perimeter compromise to actions inside an industrial control network. In another test, a model found a robotic arm, mapped its protection zones and motion limits, and moved it within minutes.
Noticeably, within these tests, existing defenses blocked or slowed some actions, but no single safeguard consistently stopped the autonomous models from achieving their objectives.
The full Booz Allen whitepaper explains, “We have long understood that non-kinetic cyber actions can produce kinetic effects. What has changed is the speed and precision with which these models can help produce those kinetic effects. They also undermine the ‘security through obscurity’ approach on which many operational technology (OT) environments have long relied.”
The Booz Allen study demonstrates what many security experts have suggested. Autonomous AI is converging cyber security and physical security. Cybersecurity alone will not control autonomous agents that will operate at a speed that human intervention cannot intercept and halt. The greater the autonomy, the more severe the threat. Autonomous agents may have the power to rewrite their own rules and take their own actions, and where OT is concerned, that can lead to potentially catastrophic effects.
The window to prepare is closing, warn the researchers. “Owners and operators should focus less on predicting every form of frontier AI use in an attack and more on making their environments difficult to traverse, removing implicit trust, detecting consequential industrial changes quickly, and limiting what an attacker can affect after gaining access, whether through cyber or physical means.”
That much is clear. What is not clear is whether this is sufficient. Enterprises are producing autonomous agents at speed and with less oversight in order to gain or maintain operational advantage over competitors. This is an economic necessity: the greater the autonomy, the more rapid the economic return. With autonomous AI, we may have unleashed a monster we simply cannot control.
And we certainly require more consistent terminology definitions to be clear about what or what aspect of AI is being discussed.
Related: Autonomous AI Hacks Raise Thorny Questions of Legal Accountability
Related: Sevii Targets AI-Speed Attacks With Preemptive Autonomous Defense
Related: OpenAI’s Upcoming Astra Model Raises Autonomous Cyberattack Concerns