Particle.news

OpenAI Buys Promptfoo to Embed Security Testing in Frontier Agent Platform

The deal brings a widely used, previously vendor‑neutral red‑teaming tool under OpenAI’s control, raising independence questions for enterprise audits.

Overview

  • OpenAI will integrate Promptfoo’s automated red‑teaming, security evaluation and risk monitoring directly into Frontier for building and operating AI coworkers.
  • The 11‑person Promptfoo team, led by Ian Webster and Michael D’Angelo, will join OpenAI to make security testing part of development workflows and compliance reporting.
  • OpenAI says it will continue Promptfoo’s open‑source CLI and library, which enterprises use to test applications and compare models across providers.
  • Financial terms were not disclosed, while PitchBook data shows Promptfoo raised about $23 million and carried a post‑money valuation near $86 million in 2025.
  • Coverage notes concern that Promptfoo may lose its vendor‑neutral standing for auditing OpenAI models even as customers gain built‑in safeguards within Frontier.