PromptArmor Collaborates with Snowflake on AI Security Research

PromptArmor is working with Snowflake to help strengthen the security of their newest AI and agentic capabilities. As an independent AI security firm, PromptArmor brings practical expertise in testing AI systems for risks such as prompt injection, unsafe tool use and failures at trust boundaries. By bringing this perspective into the development cycle, PromptArmor complements Snowflake's internal security reviews with focused, real-world adversarial testing.

Through this engagement, Snowflake may give PromptArmor scoped access to selected prerelease and preview features for structured security research. PromptArmor will test those features, privately report findings to Snowflake, and work with Snowflake's product and security teams through validation and remediation. Initial work will focus on emerging AI experiences, including the Cortex Code CLI VM sandbox (PrPr)*, with future testing chosen based on product readiness and security priorities.

This collaboration adds an external feedback loop as Snowflake expands its AI portfolio. It allows PromptArmor to help identify issues earlier, challenge security assumptions before broader release and turn independent research into actionable improvements for Snowflake customers. PromptArmor and Snowflake also plan to share appropriate lessons from the effort through joint communications, demonstrating a shared commitment to building AI features that are secure by design.

*This article contains forward-looking statements, including about our future product offerings, and are not commitments to deliver any product offerings. Actual results and offerings may differ and are subject to known and unknown risk and uncertainties. See our latest 10-Q for more information.