AI is already recognized for its potential to profoundly transform jobs in every industry. According to a new report from Citigroup researchers, "finance will be at the forefront of the changes."
"What a bank or financial firm looks like in the mid-2020s, whether in retail or wholesale finance, will differ significantly from the mid-1980s or the mid-1940s," the report states. "AI will replicate this cycle, potentially accelerating it."
While general-purpose technologies (GPTs) create new opportunities for innovation and can enhance quality of life, "they also dismantle existing methods," the report adds. "As a result, they create losers, particularly in the short term."
Drawing on data from Accenture Research and the World Economic Forum, Citi's researchers estimate that about 67% of banking jobs have a "higher potential" to be automated or augmented by AI. This implies that "banking jobs" (a term the report does not narrowly define) are at the highest risk of AI-induced job displacement.
However, Citi suggests that a reduction in headcount could be partially or entirely offset by an increase in AI-related compliance managers and ethics and governance staff.
An encouraging aspect Citi highlights is the potential profit increase. They estimate that the profit pool for the 2023 global banking sector "could rise by 9% or $170 billion due to AI adoption, increasing from just over $1.7 trillion to nearly $2 trillion."
AI adoption in finance is expected to be slow. The Citi researchers believe the "pace of implementing modern AI tools in financial services, particularly GenAI, will be relatively slow compared to other sectors," partly due to the "highly regulated nature of the sector and the lack of globally aligned rules."
"A regulatory landscape is evolving in some jurisdictions, but implementing AI in financial services remains challenging as countries move at different speeds, take varied regulatory approaches, and sometimes change their stance on whether to regulate," the report notes.
In an interview featured in the report, Shameek Kundu, head of financial services and chief strategy officer at TruEra, addresses this point.
"I would describe traditional AI adoption in financial services as: widespread, shallow, and inconsequential," Kundu says.
Kundu explains that while "a large number of enterprises are experimenting with AI across different use cases," there is a "limited scale of AI adoption across these use cases" and a "limited perceived impact of AI system failures on critical business operations."
He refers to a 2022 Bank of England survey, which found that "72% of firms reported using or developing machine learning applications," yet the "median number of ML applications for mainstream UK financial institutions was just 20-30" and "less than 20% of the already few AI use cases were critical to business."
