Cambridge, UK, Sept. 22, 2026 (GLOBE NEWSWIRE) --
- AI adoption is moving faster than security teams can keep pace with, and generative AI use is now widespread: over 80% of Darktrace's customers now use generative AI services, and in August the average organization interacted with five different AI providers.
- Early adopters of Darktrace / SECURE AI pointed to shadow AI detection, policy management, and prompt analysis as the use cases delivering immediate business value.
- The general availability of Darktrace / SECURE AI includes integrations with Amazon Web Services (AWS), Anthropic, Microsoft and OpenAI, among others, to support prompt analysis and shadow AI detection across enterprise AI platforms.
Darktrace, a global leader in behavioral security, today announced the general availability of Darktrace / SECURE AI™, giving enterprises the visibility, contextual understanding, governance and response capabilities needed to securely scale AI adoption. The solution is designed to deploy easily and rapidly, and enables organizations to discover AI use, understand how people and agents are behaving in real time, assess whether activity remains aligned with policy and intended purpose, and take action when risks emerge.
Enterprise AI adoption is now both widespread and fragmented. Over 80% of Darktrace’s customers now use generative AI services, and the average organization interacted with five different AI providers in August 2026 [1].
Early adopter deployments showed how quickly AI activity moves beyond what security teams expect, across tools, users and business use cases.
At one organization that had approved the use of a single AI assistant, Darktrace / SECURE AI discovered that a majority of employees were using unauthorized AI services, exposing a significant gap between policy and practice.
At another organization, Darktrace / SECURE AI uncovered unmanaged contractor use of multiple AI platforms, revealing a previously unknown risk vector. This prompted an immediate legal review and the implementation of formal AI governance controls.
Another organization identified nearly 90 AI agents operating within a low-code development environment with no formal approval process or oversight for agent creation, highlighting the rapid emergence of shadow AI across the enterprise.
As AI agents gain access to identities, data, applications and infrastructure, the challenge goes beyond uncovering AI use: security teams also need to know whether that behavior remains aligned with policy, intended purpose, and the normal operating pattern of the organization.
Darktrace was built for this shift. Its proprietary Adaptive AI™ learns what is normal for each organization and continuously evaluates behavior in that unique context, allowing security teams to detect emerging risk even when the activity has never been seen before. Darktrace / SECURE AI extends this behavioral approach to prompts, users, agents and AI systems, combining real-time visibility, policy governance and the ability to act when risk emerges so enterprises can move AI from experimentation to production securely and with confidence.
“Darktrace / SECURE AI has given us a clearer view of AI activity across the organization, including the tools employees are using and the types of data appearing in prompts,” said Matthew Davis, Chief Information Security Officer at ESL Federal Credit Union. “That visibility helps us understand where use falls outside our approved services, guide employees toward the right tools and make more informed decisions about licensing and education as adoption grows.”
“As we worked to develop our AI usage policy, we realized we were making decisions without a clear understanding of how employees were actually using these tools,” said David Green, Senior Vice President of IT at Lamons. “Darktrace / SECURE AI showed us that AI was already helping people work more effectively across the business, and that insight changed our approach. We can now shape policy around real behavior, protecting company data while giving employees room to use AI productively. That has given us the confidence to keep moving quickly on AI.”
“Working with cybersecurity partners like Darktrace is critical to translating advances in AI into stronger, practical defenses. Together, we are supporting organizations as they adopt AI securely and giving defenders new ways to understand and respond to risk,” said McCall McIntyre, Head of Global Cyber Partnerships, OpenAI
Behavioral Defense for Enterprise AI Adoption
Darktrace / SECURE AI gives security teams a broad, integrated approach to the safe adoption of enterprise AI:
- Shadow AI Management: Identifies unsanctioned AI activity through Darktrace telemetry and SASE integrations including Microsoft Entra Global Secure Access with the ability to block unsanctioned AI usage or quarantine affected devices. It helps customers distinguish unauthorized usage from approved usage, eliminate blind spots, and improve policy to reduce data exposure and risk.
- AI Prompt Analysis: Provides real-time visibility into prompts, sessions and responses across supported platforms, including Amazon Bedrock, ChatGPT Enterprise, Claude, Microsoft Copilot, Copilot Studio, with Salesforce support coming soon. It detects activity including attempted jailbreaks, sensitive data exposure and potential indirect prompt injection, while ranking sessions by risk to help analysts prioritize investigation.
- Policy Management: Provides a free-form policy upload capability, which enables security teams to define organization-specific controls, evaluate policy effectiveness, and continuously refine governance beyond standard frameworks. When uploading a policy, administrators can review the enterprise policy against regulatory and compliance frameworks to help users track alignment.
- AI Agent Identities and Actions: Identifies agents and human users associated with AI activity across supported environments, providing visibility into their permissions, roles, access and relationships. This helps organizations understand how agents are being used, what they are connected to and whether their activity remains aligned with their intended purpose.
- AI Agent Development Risk Management: Extends visibility into low-code and high-code development environments and platforms, including Amazon Bedrock and Microsoft Copilot Studio. Security teams can identify agent identities, excessive permissions, misconfigurations and anomalous development activity, connecting how agents are built with how they behave once deployed.
Darktrace / SECURE AI is part of the Darktrace Behavioral Defense Platform™. Rather than relying primarily on static rules, signatures, or known indicators, the platform learns what is normal for each organization and builds a real-time understanding of behavior across each enterprise. It provides unified visibility, continuous behavioral monitoring, and autonomous response across AI, people, and infrastructure, helping security teams detect suspicious activity and contain it before it causes harm.
"The security industry spent the last twenty years cataloging known threats. AI breaks that model," said Ed Jennings, President and CEO of Darktrace. "As AI systems become more adaptive and autonomous, security must understand behavior as it unfolds. Darktrace was built around a behavioral approach: you won’t always know what the next attack looks like, but you can tell when something starts behaving differently from what is normal for the organization. Darktrace / SECURE AI applies that same approach to AI, helping security teams manage risk without slowing adoption. That’s what it means to secure what AI can do.”
Advancing Behavioral Security Across the AI Landscape
Darktrace integrates with Amazon Web Services (AWS), Anthropic, Microsoft and OpenAI to help secure the rapid advances reshaping enterprise AI. Through integrations between Darktrace / SECURE AI and the frontier models, agent frameworks, and development environments customers are adopting, these collaborations extend behavioral security across agents, low-code and high-code development, prompts and enterprise AI use.
Amazon Web Service (AWS): Darktrace extends behavioral visibility across the development and runtime activity of agents built with Amazon Bedrock, helping mutual customers connect how agents are built with how they behave after deployment.
Anthropic: Darktrace extends prompt security coverage to Claude, helping mutual customers evaluate prompt and response activity in the context of each organization and identify behavior that may conflict with policy or intended use.
Microsoft: Darktrace extends behavioral understanding across Microsoft Copilot users and agents, including low-code development in Microsoft Copilot Studio. This gives customers additional context on how AI and agents behave across the wider enterprise, whether their activity remains aligned with policy and intended purpose, and when emerging risk requires action.
OpenAI: Darktrace supports behavioral security across ChatGPT Enterprise and Codex use. Through the Daybreak Defense Network, Darktrace is developing capabilities that combine OpenAI’s Daybreak models with its behavioral understanding to identify security incidents involving enterprise AI and enrich cyber incidents with business context, helping defenders understand which users, systems and business processes are affected and prioritize their response.
Availability
Darktrace / SECURE AI is available now to new and existing Darktrace customers and can be purchased as:
- A standalone product or as part of the Darktrace Behavioral Defense Platform;
- Through the Amazon Web Services (AWS) Marketplace; or
- Through the Microsoft Marketplace.
Additional Resources:
- Watch the replay of the launch broadcast, “Behavioral Security in the Age of AI”
- Register to join Darktrace LIVE in a city near you.
About Darktrace:
Darktrace secures the modern enterprise by protecting AI, people, and infrastructure with behavioral security. Founded in 2013, Darktrace uses Adaptive AI to understand what is normal for an organization and detect known, unknown and novel threats and respond autonomously in real time. The Darktrace Behavioral Defense Platform delivers unified visibility, continuous behavioral monitoring, and autonomous response across the enterprise. Darktrace protects nearly 10,000 customers across major industries globally, helping organizations defend against AI-powered threats across AI and agents, email and collaboration tools, and hybrid networks, while enabling them to innovate with AI securely.
References:
[1] Based on aggregated Darktrace product telemetry across a fleet of ~8,200 observed deployments, measuring generative AI service usage across accounts monitored in August 2026. The 80%+ figure reflects the share of monitored accounts with any generative AI service usage during the same period.
Contact Info
Darktrace Media Relations
media@darktrace.com
+1 929-316-4384
