Wall Street workers have been waiting to see how A.I. will have an impact on their work. The response from America’s fourth-largest bank has become a little clearer.
Citi Co-Chief Information Officer Shadman Zafar ushers in a new era of AI at the bank. But it will not be a project shaped by a single transformation project. Instead, he’s betting that small, incremental and sometimes “boring” steps will have the biggest impact on the bank’s overall effectiveness.
“Let’s get out of the lab and into the factory. Let everyone use it for little things,” Zafar told Business Insider. “It’s about investing it in everything you do to make a big cumulative impact.”
AI is poised to change the work of almost every Citi employee and “sustainably change the way we work for decades to come,” Zafar said. But the process will be long and will extend over several years. Zafar opens his playbook on AI, already in progress at the bank. The four-part plan will impact everyone from technology and operations to asset and product design.
Citi’s use of AI, which comes from increasing productivity, simplifying functions and automating inefficiencies, could have significant benefits for the bank. Citi has has undergone considerable changes as CEO, Jane Fraser seeks to streamline the bank’s operations and increase returns.
“We started massively deploying AI” since January, Zafar said, adding that it was too early to see the impact of the technology on the bottom line. “We focus on productivity first.”
Back office first
Naturally, Citi’s software engineers will be the first group to use genAI. This group understands the technology best and is expected to adopt it most readily, Zafar said.
The bank already has around 30,000 people using AI for software engineering, from generating new code snippets to documenting what different pieces of code do. When tens of thousands of programmers contribute code, it’s essential to know what these systems do and how they work. Zafar said AI can read, conceptualize and summarize exactly that, which can be particularly useful in increasing new hires.
“When you hire college students, you have ready-made manuals for the most complex systems, which you didn’t have before. This is actually probably the least celebrated use case , but probably the most valuable, in engineering,” Zafar said.
AI is also helping Citi modernize some of its legacy systems, a term used to describe legacy platforms that support critical back-end functions, such as loan and deposit processing and account opening . Existing systems are often written in old coding languages, like COBOL, which have become obsolete. While these legacy systems may be some of the most important, the legacy architecture is far less flexible than modern systems, making it difficult for banks to deploy new services or make changes quickly.
“With AI, you can get a complete translation of that system into a modern environment. Sometimes these are very massive projects and they become significantly simpler,” Zafar said of working with existing systems.
The next group at Citi poised to widely use AI is operations, or employees who manage back-office administrative processes. This includes everything from clearing, settling and reconciling transactions, to verifying and auditing documents and data for loan processing.
Many operational roles require repetitive, rote tasks, like sifting through documents, validating data and summarizing information, which are prime targets that AI can automate, Zafar said. Citi declined to disclose the number of people working in the bank’s operations.
AI will also change customer-facing jobs
AI will also change the work of wealth managers, private bankers and other client-facing roles, Zafar said. As the third group of workers impacted by technology, Zafar said Citi will use AI to remove “a significant amount of mechanical tasks” from their daily activities. Instead, he said bankers and advisors can spend more time developing relationships with their clients.
“Putting together different pieces of information so you can give a report to the client” is “a big part” of a banker’s job, Zafar said. Automating this process will mean bankers and advisors spend more time analyzing data, he said.
Ultimately, the final step in Citi’s AI journey will be about how the technology shapes its products, impacting everyone from product design to commercialization.
“In our industry, there is a big gap between when a customer needs it, when they want something, and when they have something,” Zafar said. While it’s not something Citi is working on today, the bank wants to use AI to deliver financial products faster and where customers spend their time, “which is not the case today.” today,” Zafar said.
