BNY Runs 125+ AI Use Cases in Production on OpenAI-Powered Platform

BNY has more than 125 AI use cases in production and 20,000 employees actively building agents — a scale of enterprise AI deployment few banks of its systemic weight have attempted.

The Bank of New York, which manages, moves, and safeguards assets across more than 100 markets and holds $57.8 trillion in assets under custody and/or administration, took its bet on generative AI when ChatGPT launched in late 2022. Rather than confine experimentation to a handful of technologists, the firm built a centralized AI Hub and an internal deployment and education platform called Eliza, then trained its workforce on responsible AI use.

"Our mantra is 'AI for everyone, everywhere, and in everything,'" says Sarthak Pattanaik, Chief Data and AI Officer at BNY. "This technology is too transformative, and we decided to take a platform-based approach for execution."

The stakes are unusually high here. BNY is one of the world's largest financial institutions and, as a systemically important bank, cannot treat AI as a side experiment. The firm's answer was to pair its governance apparatus with frontier models — including OpenAI's — inside a single controlled environment, and to push that environment out to the entire employee base rather than a privileged few.

"We are much like the circulatory system of the global financial services ecosystem," Pattanaik says. "And from that perspective, we must ensure trust is built into everything we do."

Governance as an accelerant, not a brake

The structural decision that distinguishes BNY's rollout is the integration of governance directly into the tooling. Within Eliza, all prompting, agent development, model selection, and sharing happens inside a governed environment with standardized permissions, security, and oversight.

"Some might see AI governance as a barrier, but in our experience, it's been an enabler," says Watt Wanapha, Deputy General Counsel and Chief Technology Counsel. "Good governance has allowed us to move much more quickly."

Three cross-disciplinary bodies review AI work at the bank. A data use review board brings together leaders from intellectual property, cybersecurity, engineering, data, privacy, and third-party relationships. An Artificial Intelligence release board re-examines initiatives before they reach production. The Enterprise AI Council provides senior oversight and policy alignment. Insight from the review board flows daily to the council, which evaluates high-impact or novel scenarios.

"We had to iterate as we went along," Wanapha notes. "As our use cases expand, and as the models shift, we have to constantly evaluate AI projects to maintain accuracy."

"Eliza embeds governance at the system level," Wanapha explains. "It standardizes permissions, security, and oversight across all models and tools, ensuring every workflow meets the same level of protection."

BNY chose not to build generative AI-specific governance from scratch. It extended its mature legal and compliance processes to cover new use cases, which shortened the path from idea to production.

Training 99% of the workforce

Access to Eliza is gated behind mandatory training. Every employee completes it before using the platform, reinforced by additional trainings, tools, challenges, and community support. The result: 99% of BNY's workforce is trained on generative AI.

"We introduced a number of different learning solutions to meet people where they are and to bring them along on the journey," says Michelle O'Reilly, Global Head of Talent.

One initiative stands out for its measured outcome. Make AI a Habit Month ran daily seven-minute trainings on prompting, agent building, and peer sharing. "From this month, we saw a 46% increase in the number of agents people were building," O'Reilly notes.

The cultural effects have spread beyond the platform itself. Teams from Legal, Sales, and Engineering now build side by side at bank-wide hackathons. "We had a recent hackathon in Sales," says Ed Fandrey, Head of Sales and Relationship Management. "There were no IT or tech folks present, but everyone felt like a developer."

"Before, collaboration meant more meetings," O'Reilly says. "Today, it means experimenting together, sharing prompts, testing agents, and learning by doing."

Measured results from the first agents

The first wave of agents built in Eliza delivered concrete, quantified savings. A Contract Review Assistant cuts legal review time by 75%, from four hours to one, across more than 3,000 annual vendor agreements. A People Business Partner Agent answers benefits and policy questions, reducing manual requests while improving consistency and accuracy.

Production now spans every major business line. A Lead Recommendation Engine generates client-relevant insights and opportunities. A Metrics Agent summarizes learning platform usage and performance with permission-aware access. A Risk Insights Agent uses deep research to surface emerging risk signals across portfolios, helping analysts act before issues escalate.

Eliza was built for controlled autonomy. It initially allowed only private agent builds; now, agents created by certain teams and roles can be shared with up to ten colleagues, driving reuse — one team's agent often becomes another's foundation.

Digital employees with identities and access controls

BNY has gone a step further than most enterprises by introducing advanced agents it calls "digital employees" — AI workers with identities, access controls, and dedicated workflows. They handle tasks from payment instruction validation to code security enhancements.

"Now, instead of handling certain tasks in the first instance, the role of the human operator is to be the trainer or the nurturer of the digital employee," Pattanaik says.

A select group at the bank is also experimenting with ChatGPT Enterprise, using capabilities like deep research — multi-step reasoning across internal and external data — for risk modeling, scenario planning, and strategic decision-making.

"I use it daily," says Wanapha. "If I'm tackling a novel legal question, I use deep research as my thought partner to help me evaluate whether there are questions I'm not asking."

For client-facing teams, deep research is reshaping preparation for conversations and strategic planning. Paired with agents, those insights could trigger follow-ups, draft outreach, or schedule next steps directly within client systems. Together with Eliza's orchestrator layer, these capabilities form the foundation for autonomous digital employees with permissioning, oversight, and telemetry at the core.

Why other enterprises are watching

BNY's approach offers a working blueprint for AI deployment in regulated industries: extend existing risk frameworks rather than inventing new ones, distribute review across cross-functional councils, embed governance in the interface through tagging, telemetry, approval flows, and access controls, and train the workforce before granting access.

"Unless you already know how the AI and how the platform works, you're not going to be able to really think about the risks and also the possibilities," Wanapha notes.

The bank also frames partnership as a precondition, not a convenience. "With AI, we are all encountering new questions that have not been answered," says Wanapha. "So it's very important to have the right partner and an open channel of communication."

"It's a great mix," Pattanaik says, "of the research OpenAI provides and the purposeful business case BNY provides."

The next phase is already in view. "We continue to mature beyond knowledge extraction and reasoning," says Pattanaik. "It's about connecting the dots across the organization to innovate on new products, personalized for our clients."