Workspace agents in ChatGPT: Third-party risk management agent

Third-party risk management is a compliance function that enterprises currently staff with dedicated vendor risk teams and point solutions like Prevalent or ProcessUnity. OpenAI entering this workflow, even at a demo stage, puts it in direct contact with a category that has entrenched software incumbents and meaningful regulatory liability attached to errors.

Taken alongside the same-day releases covered here, including the software review agent and the weekly metrics reporting agent, this is clearly a coordinated rollout of enterprise-facing use cases rather than isolated feature work. OpenAI appears to be stress-testing how far Codex-backed agents can reach into structured business processes before hitting reliability or trust limits. The GPT-5.5 coverage from the same date is also relevant: a 56% reduction in token consumption matters a lot if these agents are running multi-step compliance checks at scale across large vendor lists, since cost per workflow run becomes a real procurement consideration.

Watch whether any mid-market or enterprise compliance software vendor publicly responds with a partnership announcement or a competitive counter-positioning within the next 60 days. That would confirm OpenAI's workspace agent push is being read as a genuine threat to the category, not just a demo.