What Happened
OpenAI announced today it is acquiring Promptfoo, an AI security platform that helps enterprises identify and remediate vulnerabilities in AI systems during development. Once the acquisition is finalized, Promptfoo’s technology will be integrated directly into OpenAI Frontier — OpenAI’s platform for building and operating AI coworkers.
What is Promptfoo?
Founded in 2024, Promptfoo began as an open-source framework for evaluating AI prompts and model behavior. It later expanded into a commercial platform used by developers and enterprise security teams to evaluate and test applications built on large language models. The platform addresses risks including prompt injection, data leakage, jailbreak attacks, and unsafe tool execution — through an architecture that allows teams to systematically evaluate how AI systems respond to structured inputs, adversarial prompts, and real-world usage scenarios.
Despite significant market penetration, the startup remained lean — having raised only $23 million in venture capital and reaching an $86 million valuation after its most recent funding round in July 2025. Over 25% of Fortune 500 companies already use its products to stress-test their AI implementations. Investors included Insight Partners and Andreessen Horowitz.
What Gets Integrated
OpenAI will build on two core capabilities for enterprises building agents on Frontier: first, automated security testing and red-teaming capabilities will become a native part of the Frontier platform — helping enterprises identify and remediate risks like prompt injections, jailbreaks, data leaks, tool misuse, and out-of-policy agent behaviors. Second, Frontier will deeply integrate with the workflows needed to identify, investigate, and remediate agent risks earlier, making security a core part of how enterprise AI systems are developed and operated.
OpenAI also said it would continue building Promptfoo’s popular open-source project that lets developers test various AI-related prompts and agents and compare the performance of large language models like ChatGPT, Claude, and Gemini.
Why This Matters
The development of independent AI agents that perform digital tasks has generated excitement about productivity gains — but it has also given bad actors fresh opportunities to access sensitive data or manipulate automated systems. This deal underscores how frontier labs are scrambling to prove their technology can be used safely in critical business operations.
In plain terms: as OpenAI pushes Frontier deeper into enterprise environments with agentic AI, it now owns the dominant open-source toolchain used to red-team and evaluate those exact systems. It’s a vertical integration play — OpenAI builds the agents, OpenAI now also owns the security testing layer for those agents.
The Broader Acquisition Pattern
OpenAI has been acquiring startups and tech executives in recent months amid the hyper-competitive AI market. In January, OpenAI acquired the healthcare tech startup Torch for roughly $60 million. That followed its acquisition of Software Applications (AI interface “Sky” for Mac users) in October, and the hiring of Peter Steinberger who created the OpenClaw agentic development tool in February 2026.
Promptfoo fits the same thesis: every acquisition has been additive to OpenAI’s enterprise and agentic stack. Security testing was the last major gap in that stack — it’s now closed.
