OpenAI to Acquire Promptfoo to Strengthen AI Security and Testing

The acquisition brings advanced red-teaming and vulnerability-testing tools into OpenAI’s ecosystem as the company expands its enterprise AI agent platform.

Key Highlights

• OpenAI has agreed to acquire Promptfoo, a startup focused on AI security testing and red-teaming tools.
• The deal aims to improve safety and reliability of AI agents deployed in enterprise environments.
• Promptfoo specializes in detecting vulnerabilities such as prompt injection, data leakage and insecure tool use.
• Financial terms of the acquisition were not publicly disclosed.
• The move reflects OpenAI’s strategy to strengthen trust and security as AI agents become more autonomous.

OpenAI Moves to Secure the AI Agent Era

OpenAI has announced plans to acquire Promptfoo, a San-Francisco–based startup known for its tools that stress-test artificial intelligence systems and identify vulnerabilities before deployment. The acquisition is part of OpenAI’s broader effort to improve the safety and reliability of its AI platforms as it pushes deeper into enterprise applications and autonomous AI agents.

Promptfoo’s technology focuses on AI red-teaming a process where systems are deliberately tested against adversarial inputs and security risks to uncover weaknesses before malicious actors can exploit them.

Why Promptfoo Matters for AI Security

Founded in 2024, Promptfoo quickly gained traction among developers and enterprises building generative-AI applications. Its open-source framework allows teams to evaluate prompts, test model behaviour and simulate attacks such as prompt injection or data leakage.

The platform integrates automated testing into development pipelines so vulnerabilities can be identified and fixed early in the deployment process.

More than 100,000 developers and dozens of Fortune 500 companies have used Promptfoo’s tools to secure large language model applications and AI-powered services.

Strengthening OpenAI’s Enterprise Push

The acquisition comes as OpenAI expands its AI-agent ecosystem, where systems can autonomously complete multi-step tasks across digital platforms. Ensuring these agents behave safely and securely has become a top priority for enterprises adopting generative AI.

By integrating Promptfoo’s testing capabilities, OpenAI aims to identify weaknesses in AI agents at scale, allowing developers to stress-test models before deployment and prevent security failures.

Industry analysts note that as AI systems gain more autonomy interacting with software, databases and real-world systems the need for robust security and evaluation frameworks is becoming critical.

A Growing Focus on AI Safety

The acquisition reflects a broader shift across the technology sector toward AI safety infrastructure. Companies building large language models are increasingly investing in tools that monitor model behaviour, detect misuse and protect against cyber threats.

For OpenAI, acquiring Promptfoo could help reinforce trust among enterprise clients by demonstrating that security testing and responsible deployment remain central to its AI strategy.