Key Takeaway
The OpenAI Promptfoo deal would bring Promptfoo’s testing, red-teaming, monitoring, and governance tools into Frontier, OpenAI’s platform for AI coworkers. The move shows how security is becoming a core requirement as enterprises deploy more AI agents into real workflows.
The OpenAI Promptfoo dea – Key Points
The Story
OpenAI announced that it is acquiring Promptfoo, an AI security startup founded in 2024 that helps enterprises identify and remediate vulnerabilities in AI systems during development. Under the OpenAI Promptfoo deal, Promptfoo’s technology will be integrated into Frontier, while the Promptfoo team joins OpenAI. The companies said they will also continue building Promptfoo’s open-source project for evaluating and red-teaming LLM applications. The acquisition has not yet closed and remains subject to customary closing conditions.
The Facts
OpenAI said Promptfoo will be folded into Frontier.
OpenAI said Promptfoo’s technology will be integrated directly into OpenAI Frontier, the company’s platform for building and operating AI coworkers.
OpenAI alson said the integrated platform will help organizations test for prompt injections, jailbreaks, data leaks, tool misuse, and out-of-policy agent behavior.
The company said automated security testing and red-teaming will become built-in capabilities. It also said Frontier will integrate evaluation and security earlier into development workflows for agentic systems.
OpenAI said reporting and traceability features will help organizations document testing, monitor changes over time, and support governance, risk, and compliance requirements.
Promptfoo will continue building it’s open-source CLI and library for evaluating and red-teaming LLM applications.
The Promptfoo team will join the company. The company specifically named Ian Webster and Michael D’Angelo.
The acquisition terms were not disclosed.
OpenAI did not disclose the financial terms of the deal.
Promptfoo focuses on AI system security and testing.
OpenAI described Promptfoo as an AI security platform that helps enterprises identify and remediate vulnerabilities in AI systems during development. Promptfoo is a startup founded in 2024 “to protect LLMs from online adversaries”.
OpenAI said Promptfoo’s tools are trusted by more than 25% of Fortune 500 companies.
Promptfoo raised fresh capital less than a year before the deal announcement.
Promptfoo announced an $18.4 million Series A round in July led by Insight Partners, with Andreessen Horowitz also participating.
Numbers that Matter
- 25%+ of Fortune 500 companies used Promptfoo’s tools, according to OpenAI.
- About $23 million total funding raised, based on PitchBook figures cited in the reporting.
- About $86 million post-valuation as of July 2025.
Timeline / What Changed
The OpenAI Promptfoo deal adds to a recent run of acquisitions and hires tied to AI agents and enterprise infrastructure. In October, OpenAI announced it had acquired Software Applications, maker of the Sky AI interface for Mac users. In January, it acquired health-care tech startup Torch for roughly $60 million, and in February it hired Peter Steinberger, creator of the OpenClaw tool used by developers to create AI agents. This move extends that push into AI security and evaluation.
Use Cases
The integration is aimed at enterprises deploying AI coworkers into real workflows where models interact with data, tools, and internal systems. In that setting, automated red-teaming, security evaluation of agentic workflows, risk monitoring, and compliance documentation become part of deployment rather than a separate review layer.
Risks / Limitations
The acquisition is not final yet. OpenAI said the closing is still subject to customary closing conditions. Also, the companies did not disclose deal terms, so the exact financial scale of the transaction remains unknown.
Why This Matters
The OpenAI Promptfoo deal shows that AI security is moving closer to the center of enterprise AI deployment. As companies push AI agents into production systems, testing, red-teaming, monitoring, traceability, and compliance are becoming core product requirements rather than secondary tooling.
This article was drafted with the assistance of generative AI. All facts and details were reviewed and confirmed by an editor prior to publication.
OpenAI signs Pentagon AI deal for cloud-only classified deployments with three redlines, as Trump orders a six-month federal phaseout of Anthropic tools.
OpenAI will buy up to 750 MW of compute from Cerebras through 2028 in a deal worth over $10 billion to scale real-time AI inference.
OpenAI begins rolling out ads in ChatGPT for U.S. Free and Go users, with opt-out controls, clear labeling, and privacy safeguards while paid tiers stay ad-free.
Microsoft and OpenAI signed a non-binding MOU while OpenAI’s nonprofit parent gains a $100B equity stake. Regulators are reviewing restructuring and access terms.
Read a comprehensive monthly roundup of the latest AI news!
