OpenAI is buying Promptfoo. The acquisition, announced March 9, 2026, brings an AI security and evaluation platform directly into OpenAI’s enterprise infrastructure. Terms weren’t disclosed.
The strategic logic is direct. Promptfoo’s technology will be integrated into OpenAI Frontier, the company’s enterprise platform for building and operating AI coworkers. The tool already has deep market penetration: over 25% of Fortune 500 companies rely on it, and hundreds of thousands of developers have used the platform. Promptfoo was founded in 2024. That’s a two-year-old company with Fortune 500 reach at a scale most security vendors spend a decade building.
What Promptfoo does, in broad terms: it helps enterprises find and fix vulnerabilities in AI systems. According to Futurum Group’s analysis, its capabilities include prompt injection detection, data leakage prevention, jailbreak identification, and compliance monitoring, though that specific breakdown comes from a single analyst source. What’s confirmed across multiple outlets is the general characterization: an AI security platform built for identifying and remediating AI system vulnerabilities.
One detail enterprise teams will want to note: Promptfoo will remain open source following the acquisition. That commitment, made by Promptfoo directly, matters for the significant installed base of teams that built security workflows around the independent tool.
The market signal here isn’t the product. It’s the timing. OpenAI is asking enterprises to trust Frontier with mission-critical AI deployments while simultaneously acquiring the leading independent tool used to audit those deployments. That’s vertical integration of trust infrastructure. Whether it simplifies enterprise security decisions or complicates them depends entirely on how OpenAI governs the open-source commitment going forward.