Cybersecurity firm Bitdefender has introduced a new solution called GravityZone AI Visibility and Control (AIVC) to help organizations identify the AI tools used by employees and manage the resulting security risks. Officially announced on September 23, 2026, it is currently available to eligible GravityZone customers.
In many workplaces today, employees rely on far more than online AI assistants like ChatGPT; they also use coding assistants, browser/IDE extensions, locally run AI models, AI agents, and Model Context Protocol (MCP) clients/servers for their daily tasks. However, these tools are not always approved by the company's IT team. Using AI tools without authorization or clear visibility from the security team is commonly referred to as Shadow AI.
According to Bitdefender's 2026 Cybersecurity Assessment—which surveyed 1,200 IT and cybersecurity professionals across six countries—47.4% revealed that they lack full visibility into the Shadow AI tools and personal AI accounts used by employees for work.
GravityZone AI Visibility and Control addresses this challenge by detecting AI technologies across a company's managed Windows endpoints. It provides essential details for each finding—such as a risk score, category, the associated AI service, and affected devices—enabling IT administrators to prioritize and investigate high-risk activities first.
For example, if an employee uploads a confidential document to an unknown public AI chatbot instead of a company-approved platform, inputs source code into an unauthorized coding assistant, or runs an unapproved local AI model, AIVC helps the organization detect that AI usage and evaluate the associated risk.
There is also an important privacy aspect to this solution: Bitdefender notes that AIVC does not read prompts, AI chat histories, passwords/credentials, model weights, or full file contents. Instead, it collects only the minimal metadata necessary from the endpoint to identify AI usage and evaluate security risks.
Once a risk is detected, administrators can leverage GravityZone’s existing web filtering, application control, and endpoint hardening controls to apply policies that allow, monitor, restrict, or block specific AI services. However, the current release of AIVC does not automatically block or remediate findings on its own; its primary goal is to provide the security team with clear visibility and mitigation guidance.
With the rapid adoption of generative AI tools in the workplace, the risk of inadvertently sharing confidential company data, customer information, internal documents, or proprietary source code with third-party AI systems is constantly growing. If an IT department does not know which AI tools are being used within the organization, enforcing a security policy becomes nearly impossible.
Through Bitdefender GravityZone AI Visibility and Control, organizations can close this Shadow AI visibility gap. It allows IT security teams to move beyond asking "Are employees using AI?" to answering "Which AI tools are being used, where are they being used, and which ones pose the greatest risk?"
