- At Money 20/20 in Amsterdam this week, top executives from the banking industry and fintech gave a number of explanations for generative AI, from “mind-boggling” to “an explosion of innovation.” Used terms and phrases.
- Automate complex banking processes with generative AI that generates content in response to user prompts.
- But the technology is still in its infancy, and many banks warn that it may be too risky to adopt in consumer-facing areas.
The GPT-4 logo can be seen in this photographic illustration taken in Warsaw, Poland on March 13, 2023.
Jaap Ariens | Null Photo | Getty Images
AMSTERDAM, NETHERLANDS — Big banks and fintechs claim they’re rushing into generative artificial intelligence at a time when the hype around hot tech shows no sign of dying off, but potential pitfalls and risks. concerns persist.
At the Money 20/20 fintech conference in Amsterdam, the Netherlands, executives from leading financial institutions and online finance companies praised generative AI, calling it an “innovation explosion” and saying, “We We will unleash innovation wherever possible.” I don’t even think about it. ”
Chalapathy Neti, head of AI at global banking messaging network Swift, calls the progress of ChatGPT and GPT-4 “mind-boggling”. “This is truly a moment of change,” he added.
But in the short term, banks are scrambling to find use cases.
ABN Amro in the Netherlands is one of the major banks piloting the use of generative AI in their processes.
ABN Amro’s Chief Commercial Officer for Personal and Corporate Banking, Annely Vroigdenhill, said in a panel discussion that the company is using technology to automatically summarize conversations between bank employees and customers. We also use it to help our employees collect data about customers, help answer questions, and avoid repetitive questions.
The bank is currently in the process of expanding these pilots to 200 employees and is considering a number of new pilots to launch this summer.
Meanwhile, in a private session on the application of AI in financial services, two bank executives described how they are using technology to improve internal code and analyze customer behavior.
“At this stage, we are still experimenting and not necessarily what our clients will face. [tech the] For example, code refactoring and telephony and vice versa, just like any other company,” said Mariana Gómez de la Villa, an executive at ING Bank specializing in strategy and innovation.
In fact, banks seemed unanimously hesitant to deploy tools like ChatGPT in customer-facing scenarios.
Jon Ander Veracochea Alaba, head of advanced analytics discipline at Spanish bank BBVA, said the bank has a “conservative approach” to AI, saying generative AI is currently ” It’s still in its early stages,” he added, adding that it’s “immature.”
The key issue is that advanced AI systems need to process vast amounts of data, highly sensitive data wrapped in all sorts of rules and regulations. As a result, Mr. Alaba said it was “too risky” to handle confidential information from customers at this stage.
Generative AI is a special form of AI that can create content from scratch. The system takes input from users and feeds it into powerful algorithms leveraging large datasets to generate new text, images and videos in a more human way than most AI tools already on the market. increase.
The technology came into the spotlight following the success of OpenAI’s GPT language processing technology. ChatGPT, which uses a large language model to create human-like responses to questions, has sparked an arms race among some companies over what is seen as the next “paradigm shift” in technology. .
Goldman Sachs chief information officer Marco Argenti told CNBC in March that the bank is experimenting with generative AI tools internally to help developers automatically generate and test code. said.
Most recently, in May, Goldman spun off its first startup from the bank’s internal incubator: an AI-powered enterprise social media company called Louisa. The AI effort is part of a larger effort by CEO David Solomon to accelerate the bank’s digital transformation.
Morgan Stanley, on the other hand, uses it to notify its financial advisors when they have inquiries. So far, the bank aims to help about 16,000 advisors leverage Morgan Stanley’s research and data repository, according to Jeff McMillan, head of analytics and data at the bank. So far, they are testing a chatbot using OpenAI with 300 advisors. The company’s wealth management department.
These are just a few examples of how financial firms are using AI, but not as a core part of their services, but as digital helpers.
Gudmundur Kristjansson, CEO and co-founder of Icelandic regulated technology company Lucinity, told CNBC how artificial intelligence can be used to help fight crime, a key area of finance. introduced.
The company has developed an AI tool called “Luci” that aims to help publicize compliance.ahAn expert who conducts research. During the live demo, Christiansson was shown investigating a money laundering case. AI tools analyzed the incident, described what they saw, and completed an independent review.
In this use case, rather than replacing the role of humans investigating reports of suspicious activity, AI acts as a resource, or “co-pilot,” to help employees find data and crystallize cases. Works.
“When we find out that money laundering is taking place, … An interconnected network of people who are basically employed to do it. That’s why it’s so hard to find. Banks have spent $274 billion on prevention this year,” Christianson said in an interview with CNBC.
He said Luci will help significantly reduce the amount of time spent figuring out whether something is a scam or money laundering.
According to Money 20/20 participants, the biggest attraction of AI for big banks and fintechs is that it will reduce the time and money it takes for human employees to complete tasks that could take days. It is said that there is a possibility that
Niclas Guske, COO of Tactile, a start-up that helps fintech companies automate decision-making, said that given the lack of publicly available data, He acknowledged that using AI is difficult.
But he stressed that it could be a “key” tool for reducing operating costs and improving efficiency for businesses.
“Many fintech applications do this by increasing automation and reducing manual processes, especially in onboarding and underwriting,” he told CNBC.
“This automation is truly enabled by accessing more data sources, allowing lenders to gain new insights and find the right information without parsing dozens of PDFs for the right information. We will be able to identify the best customers.”
— CNBC’s Hugh Son contributed coverage.
