OpenAI has reached an agreement to acquire Promptfoo, a startup that helps large companies identify and address security vulnerabilities in artificial intelligence models during their creation, as part of an initiative to assist corporate clients in minimizing potential risks associated with implementing AI software.

The creator of ChatGPT is expected to reveal on Monday that it will incorporate Promptfoo’s technology into OpenAI Frontier, a newly launched platform designed for organizations to develop and oversee AI agents. Frontier is designed to guarantee that every agent — referred to as “AI coworkers” by OpenAI — has suitable safeguards and access to data.

OpenAI and its competitors are competing to create more sophisticated AI agents capable of handling complex tasks on behalf of users, requiring minimal human involvement. Simultaneously, the AI developer aims to persuade a wider variety of companies to invest in the technology by making sure these products are both effective and secure.

"OpenAI's acquisition of Promptfoo demonstrates a strong dedication to ensuring that enterprise AI is not only potent but also secure and dependable on a large scale," stated Ganesh Bell, managing director at Insight Partners, which spearheaded an $18.4 million funding round for the security startup last year. Andreessen Horowitz took part in the round as well.

Also Read: Republic Day 2026: India's Tech Triumphs & Global Leadership

Promptfoo, a two-year-old company, creates open-source tools for evaluating the security of AI systems and assists businesses in attempting to challenge their own products to identify vulnerabilities, a practice called red-teaming. The startup located in San Francisco has around 25 percent of Fortune 500 companies as clients.

Also Read: AI & Tech: Visionary Pre-Budget Insights from Industry Leaders

Promptfoo is among numerous startups creating cybersecurity solutions that leverage AI to protect against hackers, who are increasingly using similar technologies to search for vulnerabilities in essential networks.

Also Read: NVIDIA's Rubin Moment at CES: AI Supercomputing Breakthrough

OpenAI has also taken steps to incorporate security features into its AI products and agents. The company launched an AI agent last week designed to assist security teams in identifying and fixing vulnerabilities in extensive databases, akin to a tool from competitor Anthropic PBC.