U.K.-based cybersecurity company Darktrace Holdings Ltd. today made its SECURE AI offering generally available, bringing its behavioral detection approach to corporate use of artificial intelligence tools and agents.
The release is aimed at AI use that has spread through the enterprise faster than security teams can keep track of it. Telemetry from about 8,200 Darktrace deployments shows more than 80% of monitored customer accounts used generative AI services in August, and the average organization dealt with five different AI providers that month.
The company offered three examples from early customer deployments. One organization had approved a single AI assistant, only to find most of its employees using unauthorized services. At another, nearly 90 AI agents had been built inside a low-code development environment with no approval process for creating them. Unmanaged use of several AI platforms by contractors at a third company led to an immediate legal review.
Darktrace’s security products spot threats by learning what normal activity looks like inside each customer. SECURE AI points that same detection at AI use, and the general availability release includes integrations with Amazon Web Services Inc., Anthropic PBC, Microsoft Corp. and OpenAI Group PBC.
Sessions in ChatGPT Enterprise, Claude, Microsoft Copilot and Amazon Bedrock are scanned in real time for jailbreak attempts and sensitive data exposure, along with signs of indirect prompt injection. Each session is ranked by risk so analysts know where to start.
Shadow AI detection leans partly on secure access service edge integrations such as Microsoft Entra Global Secure Access. Administrators can block unsanctioned services or quarantine a device outright. Companies can also upload their own AI policies in free-form text and have the software check them against regulatory and compliance frameworks.
Darktrace also maps AI agents to the human users behind them and records what each agent is permitted to reach. Coverage extends back into development environments such as Amazon Bedrock and Microsoft Copilot Studio, where excessive permissions and misconfigurations can be caught while an agent is still being built.
Ed Jennings, who took over as Darktrace’s president and chief executive in March, said the security industry spent two decades cataloging known threats. “AI breaks that model,” he said. The next attack will not always be recognizable in advance, in his view, so the signal that matters is when something “starts behaving differently from what is normal for the organization.”
Darktrace is a member of the Daybreak Defense Network, OpenAI’s group of cybersecurity partners. Under that program, Darktrace is developing capabilities that pair OpenAI’s Daybreak models with its own behavioral data, with the aim of showing defenders which people and systems an AI-related incident has touched. McCall McIntyre, head of global cyber partnerships at OpenAI, said working with partners such as Darktrace is “critical to translating advances in AI into stronger, practical defenses.”
Announced in February, SECURE AI is on sale now to new and existing Darktrace customers as a standalone product or as part of the Darktrace Behavioral Defense Platform. The AWS Marketplace and Microsoft Marketplace carry it as well.
