.png)
TL;DR Prompt Security’s solution for Employees now supports Claude’s recently released desktop application, delivering comprehensive protection and governance over the tool.
We're proud to announce that our lightweight GenAI Security agent now supports Claude's desktop application, further expanding our comprehensive protection for employees using AI tools.
Desktop AI applications are on the rise
While browser-based GenAI tools have dominated the landscape, we’re seeing more and more AI tools, copilots and chatbots releasing their version for desktop. Examples of these are OpenAI’s ChatGPT desktop app or Microsoft Copilot, which is built-in on Windows 11. Claude, the AI chatbot developed by Anthropic, has recently joined this trend by launching its own desktop app for Mac and Windows.
The application allows the Claude 3.5 Sonnet model to control a computer by looking at a screen, moving the cursor, clicking buttons, and entering text. This system can take screenshots, make decisions based on visual input, and execute bash commands. While groundbreaking, it also opens the door to serious risks, particularly through prompt injection attacks, potentially allowing exploitation by malicious actors. In practical terms, a malicious actor could easily install malware, perform remote shell executions, delete all stored data, exfiltrate sensitive information, and more. We explored all of these risks in depth in our recent blog post.
As part of our commitment to protect organizations from GenAI risks, regardless of where the tools are being used (browser or desktop), several months ago we launched a lightweight agent as part of our offering of Prompt for Employees. The agent delivers the same level of security and governance on desktop applications as we do on the browser. Notably, we were the first security vendor to support ChatGPT for Desktop, and more recently, Copilot for M365, among many other tools.
Today we’re happy to share that we’re extending the supported desktop applications to Claude for Desktop.
Prompt Security's Solution: Lightweight GenAI Security Agent
Prompt Security inspects each prompt and model response to prevent the exposure of sensitive data and secure against GenAI-specific attacks. With the lightweight agent, the solution can block or redact prompts containing sensitive data in order to prevent leaks via GenAI applications on desktop.
Key features of our desktop agent include:
- Real-time prevention of sensitive data exfiltration, ensuring full data privacy and compliance
- Rapid deployment: The agent can be installed on employees' computers in just minutes
- Department-based granular policies for the use of Claude for Desktop, adapting to the unique needs and risk appetite of the organization
- Comprehensive auditing of all data shared with and retrieved from Claude for Desktop
Conclusion
By extending our support to Claude's desktop application, Prompt Security reaffirms its commitment to providing comprehensive GenAI security across all platforms. As GenAI tools become increasingly integrated into daily workflows our solution ensures that organizations can embrace these technologies without compromising on security or compliance.
