Training Agents On Contracts
OpenAI announced a research collaboration with Ironclad focused on turning complex contracting workflows into training and evaluation tasks for AI agents. OpenAI reported that its research evaluation scored 32% higher than GPT-5.6 Sol’s, with estimated time per attempt 48% lower; this is a research collaboration, not a generally available product release.
OpenAI announced a research collaboration with Ironclad focused on turning complex contracting workflows into training and evaluation tasks for AI agents. OpenAI reported that its research evaluation scored 32% higher than GPT-5.6 Sol’s, with estimated time per attempt 48% lower; this is a research collaboration, not a generally available product release.
OpenAI published the announcement on October 6, 2026, within the requested 24–72-hour window.
OpenAI says its collaboration with Ironclad will turn complex contracting workflows into training and evaluation tasks for computer-use agents. The research evaluation reportedly scored 32% higher than GPT-5.6 Sol’s, with 48% less estimated time per attempt. Which contracting step would be most important to test first? martinezaistudios.com