The bank’s initial focus includes processes such as client and employee onboarding, and some know your customer (KYC) processes, according to the report.
It is already using artificial intelligence to streamline account openings, automate coding and migrate data from legacy systems. In one application of the technology, AI has reduced the time needed to review documents before opening an account in the U.S. for the services division from an hour and 15 minutes to just 15 minutes, the report said.
Citi is reducing the percentage of its technology workforce that is made up of contractors from 50% a year ago to 20%, and it has increased its tech workforce to about 50,000 people as it develops and deploys AI tools, per the report.
“We still have legacy systems, but after the tech investments we’ve made, we are in a much better place,” Ryan told Reuters.
The PYMNTS Intelligence report “2025 State of Fraud and Financial Crime in the United States” found that financial institutions are investing more aggressively in AI, behavioral analytics and cloud infrastructure to keep pace with increasingly sophisticated forms of fraud.
It was reported in February that agentic AI is moving beyond experimental pilots and into the operational core of financial institutions. These institutions are embedding agentic systems into compliance, treasury, risk and payments infrastructure.
In March, it was reported that banks are using autonomous AI tools to prepare client materials, model financial scenarios, speed outreach and improve fraud detection.
Ryan said in a post on LinkedIn that as of the end of 2025, more than 182,000 Citi employees in 84 countries were using the bank’s AI tools.
“We launched new AI products and capabilities that are changing how we work and are helping our teams build new skills,” Ryan wrote in the post.
In another post on LinkedIn, Ryan said that Citi has rolled out prompt training and that colleagues had input more than 6.5 million prompts, “transforming tasks that once took hours into minutes.”