OpenAI and Ironclad partner to train and evaluate AI agents on legal contracting workflows

OpenAI and Ironclad have announced a partnership to train and evaluate AI agents on complex contracting workflows. The collaboration is framed as an effort to advance computer-use capabilities for professional work, with a focus on legal and contracting domains. The announcement does not disclose financial terms, technical specifics, or benchmarks for the agents involved.

The initiative highlights a shift from static AI assistance to dynamic, task-oriented agents capable of navigating multi-step workflows. Ironclad, a company specializing in contract lifecycle management, will provide the domain-specific workflows for training and evaluation. OpenAI’s role appears to center on developing and refining the agentic systems that interact with these workflows.

While the announcement lacks concrete metrics or release timelines, it underscores the growing interest in deploying AI systems that can autonomously perform professional tasks. The focus on contracting workflows suggests a testbed for evaluating agent reliability, accuracy, and safety in high-stakes, real-world scenarios.