JPMorgan Chase signs OpenAI deal to use Codex and cybersecurity models

JPMorgan Chase has signed a strategic partnership with OpenAI to use its Codex coding agent and its cybersecurity models, Finextra reported. OpenAI president Greg Brockman announced the deal on stage at the WSJ Tech conference on November 4, saying "we just papered a partnership with JPMorgan Chase, and that they're going to be using Codex."

Brockman said the bank will use OpenAI's cyber models "to secure themselves" and deploy its models "for enterprise transformation knowledge work across the company." OpenAI will provide the cybersecurity models through its Daybreak program, which gives approved organisations access to advanced cybersecurity capabilities, PYMNTS reported, citing The Wall Street Journal. In a post on X, Brockman said he was "especially excited about working together on cybersecurity, to help secure and protect the financial system to a level that is only possible with AI."

Neither company disclosed financial terms, contract value or duration, MLQ reported. JPMorgan has used OpenAI models since 2024 through LLM Suite, its internal AI platform, which also offers Anthropic models, and the bank's work with Anthropic on the Project Glasswing cybersecurity effort continues alongside the new agreement.

Brockman disclosed the deal in the same appearance in which he said OpenAI could be ready to go public in 2027 and that most of its revenue now comes from enterprise customers.