DTEX adds AI risk management to track how agents and employees use AI

DTEX Systems Inc., a behavioral intelligence security company, today announced an enhanced AI risk management product that reads the intent behind how employees and autonomous artificial intelligence agents use generative AI tools across the enterprise.

DTEX claims that this release targets a gap that most security tools have not yet filled. Once co-pilots, generative AI applications, and AI agents have access to corporate data, systems, and workflows, traditional tools can log the behavior of those systems, but they cannot determine whether that behavior is consistent with intent. DTEX is touting AI risk management as a way to read intent across both humans and machines.

The product is designed to detect sanctioned and unauthorized AI use across users, endpoints, and workflows, covering browsers, applications, and embedded AI activity. Flag shadow AI and a built-in co-pilot in real-time to build an inventory of approved tools and categorize risks for unknown tools. It also monitors prompts, responses, and data movement at a granular level to discover leaks of source code, intellectual property, and other sensitive data.

The central claim is that it is possible to distinguish between human and AI actions. According to DTEX, the product provides visibility into so-called Computer Use AI, which tracks what agents were told to do, how they performed their tasks, and the lineage of actions taken across the system. The company says that by correlating prompts, behavioral baselines, and agent actions over time, it can separate routine experiments from dangerous or malicious behavior and detect agent-driven data exfiltration before it becomes a breach.

DTEX said one of its early deployments identified autonomous agents that exposed sensitive data while operating within their intended workflows and privileges, surfacing risks before an incident occurred.

“AI agents are rapidly becoming operational actors within enterprises, with the ability to access data, interact with systems, and take autonomous actions,” said CEO Marshall Heilman. “Many security solutions can monitor AI activity, but cannot determine whether the behavior is in line with intent or poses a risk.”

In addition to monitoring tools, DTEX is adding two autonomous security agents that act based on what the platform detects. Triage Guardian uses a multi-agent approach to automate investigation workflows, collect evidence, and run independent review agents that validate results and reduce false positives. Threat Hunter allows analysts to initiate investigations in natural language and correlate results to uniquely surface unknown threats.

Both leverage DTEX i3 research, including collaboration with MITER and Five Eyes defense partners. The company says early adoption saved more than 40 hours per analyst per month.

Founded in 2000, DTEX raised $50 million in 2024 from Alphabet Inc.’s growth fund CapitalG to integrate insider risk management, data loss prevention, user behavior analytics, and user activity monitoring into a single platform.

AI Risk Management is available in private preview and will be more widely available in the next quarter.

Image: DTEX