OpenAI to acquire AI security platform Promptfoo to bolster enterprise agent safety

On March 9, OpenAI announced its intention to acquire the AI security startup Promptfoo, aiming to enhance its testing and evaluation tools for enterprise-level AI systems. This move comes as more companies integrate autonomous AI agents into their operational workflows. Promptfoo's advanced technology is set to be incorporated into OpenAI Frontier, a platform tailored for developing and managing AI coworkers in business environments. While the financial specifics of the acquisition remain undisclosed, it is subject to standard closing conditions. The tools provided by Promptfoo are essential for identifying vulnerabilities in AI systems during their development phases. These include critical issues like prompt injections, data leaks, and the misuse of interconnected tools. Their platform is widely utilized by developers for testing and conducting red-team exercises on large language model applications. Srinivas Narayanan, the Chief Technology Officer of B2B Applications at OpenAI, expressed enthusiasm for the acquisition, stating, "Promptfoo brings deep engineering expertise in evaluating, securing, and testing AI systems at enterprise scale. Their work helps businesses deploy secure and reliable AI applications, and we’re excited to bring these capabilities directly into Frontier." With the integration of Promptfoo’s automated security testing and red-teaming features, OpenAI aims to empower enterprises to identify risks before launching AI agents. As companies begin to link AI systems with their internal data and operational tools, issues of governance, compliance, and oversight have surged to the forefront of generative AI adoption. OpenAI also plans to incorporate reporting and traceability functionalities into Frontier, allowing companies to document their testing processes, track system modifications over time, and adhere to regulatory standards. Promptfoo’s tools are already favored by a significant portion of developers, including an open-source command-line interface designed for evaluating and stress-testing AI applications. The company claims that over 25% of Fortune 500 companies trust its platform. Ian Webster, co-founder and CEO of Promptfoo, noted, "We started Promptfoo because developers needed a practical way to secure AI systems. As AI agents become more intertwined with real data and systems, ensuring their security and validation is increasingly crucial. Joining OpenAI enables us to expedite this work, enhancing security, safety, and governance for teams developing real-world AI systems."