OpenAI Group PBC announced today that it plans to acquire testing and security evaluation platform for artificial intelligence applications startup Promptfoo Inc. for an undisclosed price.

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 and other generative AI systems.

The company’s platform addresses risks associated with deploying AI models in production environments. Those include prompt injection, data leakage, jailbreak attacks and unsafe tool execution. It does so through an underlying architecture that allows developers and security teams to systematically evaluate how AI systems respond to structured inputs, adversarial prompts and real-world usage scenarios.

The technology works through a testing framework that allows teams to define prompts, expected outputs and evaluation criteria in configuration files. The system runs the prompts across one or more language models or AI applications to capture responses and score them against predefined rules or automated grading functions.

The framework supports integrations with common development tools and application programming interfaces to allow evaluations to be executed locally, in continuous integration pipelines, or as part of automated deployment workflows. Core functions include prompt testing, red-team simulations and evaluation dashboards that document model behavior under different conditions. The system also supports regression testing, allowing organizations to compare outputs across model versions or configuration changes to identify behavioral differences before software updates are released.

The platform is currently used in environments where organizations are building or operating AI-powered applications such as chatbots, agent frameworks and automated decision tools. OpenAI plans to integrate Promptfoo’s technology directly into OpenAI Frontier, its platform for building and operating AI coworkers. Promptfoo’s technology will also be used by OpenAI to allow enterprises to test agent behavior, detect risks before deployment and maintain clear records to support oversight, governance and accountability over time.

“Promptfoo brings deep engineering expertise in evaluating, securing and testing AI systems at enterprise scale,” Srinivas Narayanan, chief technology officer of B2B applications at OpenAI, said in an announcement post. “Their work helps businesses deploy secure and reliable AI applications and we’re excited to bring these capabilities directly into Frontier.”

Coming into its acquisition, Promptfoo had raised $23.6 million in funding, including a round of $18.4 million in July. Investors in the company include Insight Partners LP and Andreessen Horowitz.

Image: Promptfoo