Kitecyber, an endpoint-native data security company, today launched AI Workflow Security, part of its GenAI and Agent Security offering. The capability governs how users, copilots and AI agents access and act on sensitive data across business applications, directly at the endpoint, with no gateways, appliances or added infrastructure.
The launch comes as AI shifts from passive assistance to active participation in business workflows. Stanford's 2026 AI Index reports organizational AI adoption has reached 88 percent. Microsoft's 2025 Work Trend Index found that 81 percent of leaders expect agents to be moderately or extensively integrated into their AI strategy within 12 to 18 months, and 46 percent say their organizations already use agents to fully automate workflows.
As AI moves from generating responses to reading data, using tools and completing tasks, the security question changes. Organizations must govern not only who has access, but whether a user or agent should move a specific piece of data at a specific moment. That decision happens on the endpoint, where the data actually moves.
Security has not kept pace. IBM's 2025 Cost of a Data Breach Report found that 63 percent of organizations lacked AI governance policies to manage AI or prevent shadow AI, and that among organizations reporting an AI-related security incident, 97 percent lacked proper AI access controls.
Many AI security approaches focus on prompt filtering, model guardrails or network gateways. Those controls address part of the workflow. Kitecyber places the control on the data itself, at the point where a user or agent tries to read, copy, transform, upload, share or act on it across browsers, SaaS applications, local tools and endpoints. The agent applies a real-time policy decision using the context of the user, device, data, application, agent, tool and destination, then lets trusted work continue.
Kitecyber describes the model as See, Decide, Enforce, continuously. The agent observes each data action, evaluates it in context, and enforces the right control at the moment work happens.
"The prompt is one moment. The action is where risk becomes real," said Ajay Gulati, CEO and co-founder of Kitecyber. "AI agents operate at machine speed and can touch data and tools across an entire workflow. Security teams cannot rely on network-layer controls alone. They have to decide, in real time and at the point of execution, whether an action involving sensitive data should be allowed."
"Every employee is about to have a productivity agent working on their behalf, and those agents move data faster than anyone can review it," said Gulati. "The way a company adopts that technology safely is a security agent on the same endpoint, deciding what each action is allowed to do with sensitive data."
In practice, Kitecyber lets organizations discover shadow AI usage, see how data is accessed and moved, and enforce guardrails across approved and unapproved tools. The platform can stop an agent running in a browser or on an employee device from sending customer records to an unapproved AI service, while letting an authorized workflow proceed with contextual guidance and a full audit trail.
AI Workflow Security is delivered through a single lightweight endpoint agent and extends Kitecyber's existing platform, which brings together data loss prevention, endpoint management, secure web and SaaS access, and zero trust private access. Consolidating these controls on one agent removes the separate point products and added infrastructure most teams run today. The offering is available now through Kitecyber's existing endpoint agent.
About Kitecyber
Kitecyber is an endpoint-native cybersecurity company that puts data security at its source, the endpoint, where work happens. Headquartered in the San Francisco Bay Area, the company delivers data loss prevention, GenAI and AI agent security, endpoint management, secure web and SaaS access, and zero trust private access through a single lightweight agent. Kitecyber enables organizations to see data actions, make context-aware decisions, and enforce controls at the moment work happens.
Media Contact
Company Name: Kitecyber
Contact Person: Ajay Gulati
Email: Send Email
City: San Francisco
State: California
Country: United States
Website: https://www.kitecyber.com/
