Promptfoo Lands $18.4 Million

promptfoo-logo

SAN FRANCISCO &&num;8212&semi; Promptfoo, an openâsource framework for AI redâteaming and security testing, has landed an &dollar;18&period;4â¯million Seriesâ¯A financing led by Insight Partners, with participation from existing investor Andreessen Horowitz&period;&NewLine;Founded in 2024, Promptfoo is steadily gaining recognition as an emerging standard for robust AI security testing, with over 100,000 developers using its open source tools and adoption by more than 30 Fortune 500 companies&period; The platform helps developers and security teams secure large language models (LLMs) and generative AI applications by detecting and mitigating risks like prompt injection, data leakage, and insecure tool use&period; Its commercial enterprise solution, launched just a year ago, is trusted by leading retailers, telecoms, and financial institutions to securely deploy AI systems&period;&NewLine;The funding comes at a critical point for enterprise AI adoption&period; As companies move beyond proof-of-concepts to production deployments, the complexity and risk profile of AI systems has grown exponentially&period;&NewLine;&&num;8220&semi;AI security has become the largest blocker to enterprises shipping generative AI applications to end users,&&num;8221&semi; said Ian Webster, CEO and co-founder of Promptfoo&period; &&num;8220&semi;Architectures like RAG, agents, and most recently MCP have expanded what&&num;8217&semi;s possible with AI, but they&&num;8217&semi;ve dramatically increased the attack surface&period;&&num;8221&semi;&NewLine;Promptfoo&&num;8217&semi;s platform stands apart as one of the only AI-security solutions that closes the loop from discovery to resolution&period; It embeds automated red-team tests directly into the build and release process, pairs findings with clear, actionable remediation guidance, and feeds results into existing vulnerability-management workflows&period; This end-to-end automation—spanning attack objective generation, exploit search, and regression testing—gives security and engineering teams a single, continuous workflow for hardening today&&num;8217&semi;s complex agentic systems, establishing Promptfoo as a leading choice for organizations aiming to deploy AI securely at scale&period;&NewLine;&&num;8220&semi;Promptfoo has created what we believe to be a category-defining product,&&num;8221&semi; said Ganesh Bell, Managing Director at Insight Partners&period; &&num;8220&semi;The layers of The Next Stack built around AI are clearly emerging beyond LLMs, and Promptfoo is a critical part of that evolution—especially in the race to enable intelligence-first design and secure Enterprise AI&period; Serious enterprises deploying AI will need comprehensive security testing, and Promptfoo&&num;8217&semi;s combination of technical depth, open source adoption, and enterprise traction positions them perfectly to capture this massive market opportunity&period; We&&num;8217&semi;re thrilled to partner with them as they help build the full AI security suite&period;&&num;8221&semi;&NewLine;The funding will fuel Promptfoo&&num;8217&semi;s expansion as it moves to establish the industry standard for enterprise AI security&period; The company will scale its team and accelerate platform development to meet demand from Global 2000 companies seeking comprehensive solutions to secure their AI deployments&period;&NewLine;&&num;8220&semi;When we first invested in Promptfoo, we recognized that AI security would become mission-critical for every enterprise,&&num;8221&semi; said Zane Lackey, General Partner at Andreessen Horowitz&period; &&num;8220&semi;In just one year, they&&num;8217&semi;ve validated that thesis and then some&period; This additional investment reflects our conviction that they&&num;8217&semi;ll define how companies approach AI security for years to come&period;&&num;8221&semi;&NewLine;

Editor