OpenAI and Ironclad Collaborate to Enhance AI in Complex Contracting Workflows

News

OpenAI partners with Ironclad to train AI agents on complex contracting workflows, leveraging GPT-6 Astra and advancing AI’s professional application capabilities.

OpenAI has announced a significant partnership with Ironclad to train and evaluate artificial intelligence (AI) agents on complex contracting workflows. This research effort, revealed on October 6, 2026, aims to advance the use of AI in professional work environments, particularly in specialized business software applications.

Evidence and context

The collaboration focuses on the development of AI models capable of executing multi-step contracting workflows, such as agreement configuration, approvals, and document management. OpenAI’s latest model, GPT-6 Astra, was trained specifically on tasks designed by Ironclad, a company with expertise in contract lifecycle management. According to OpenAI, GPT-6 Astra outperformed its predecessor, GPT-5.6 Sol, with a 32% improvement in research evaluation scores and a 48% reduction in estimated time per task attempt.

Importantly, OpenAI emphasized that no customer data from OpenAI or Ironclad, nor any nonpublic data, was used in the training or evaluation of these models. Instead, Ironclad contributed its domain expertise to define realistic and high-value tasks for developing AI capabilities tailored to contracting workflows.

This partnership builds on a history of collaboration between the two companies. Ironclad has previously integrated OpenAI’s GPT-4 technology into tools like AI Assist, which assists legal teams in reviewing and negotiating contracts. Additionally, Ironclad’s support documentation affirms that its agreement with OpenAI includes a do-not-train clause for customer data, ensuring data privacy and security.

The broader implications of this research effort are substantial. By improving the ability of AI agents to navigate and execute complex business operations, OpenAI and Ironclad aim to enhance productivity and address challenges in professional domains where precision and compliance are critical. Earlier this year, Ironclad also announced an AI assistant aimed at bringing contract intelligence to every agreement, further underscoring its commitment to advancing AI in legal and business processes.

The collaboration reflects a growing trend of integrating AI into specialized professional workflows, with companies like Deloitte previously forming alliances with Ironclad to expand agentic contract lifecycle management capabilities. However, the timeline for broader commercialization of these advanced AI tools remains uncertain.