Page 18 - GS230802

Views

How AI may innovate ing its worth more in the provision of innovative financial
services rather than being an alternative to card payments
open banking at checkouts and ecommerce sites.

As stated by the 7 million figure above, open banking us-
age is low compared to buy now, pay later services, which
are used by 17 million people in the UK (see https://tinyurl.
com/57vy8yra). But it has continued to increase and may rise
from 15 percent of the public to 44 percent by 2027 if this
trajectory continues (see https://tinyurl.com/yy53azmc).

The rise could be attributed to open banking services being
integrated into more common processes as opposed to ma-
jor companies including Klarna. Some consumers signing
up for mortgages or loans may use open banking systems
for examining their bank account. Many consumers may
not even know that an open-banking transaction has taken
By Scott Dawson place; it would just seamlessly integrate into processes that
DECTA they already use (such as subscription payments).

pen banking has a lot of potential for growth. Can AI solve open banking challenges?
Open banking systems were used by 7 million
people and businesses in January 2023, accord- AI has been prominent in payments and open banking for
O ing to Open Banking Ltd. research, but this is quite a while, so integrating machine learning and big data
still a small percentage of the 156 million bank accounts in into financial services is not new. Even if it is just a sorting
the UK (see https://tinyurl.com/yvxvsdpz). algorithm used for basing decisions on customer account
information, newly registered AISPs will already use some
Furthermore, while Bud, Plaid and Tink have been form of machine learning. The "if X then Y" logic has been
successful on their own terms, they can’t compare to the used by most basic banking systems, such as the system
rapid growth of household names like Klarna and Revolut. that sends you a text if you are overdrawn. These could be
Open banking also has low awareness and understanding classed as a form of AI since no humans are making that
among the general public; some people even question why decision.
open banking still exists (see https://tinyurl.com/577f9ssk).
In general, AI refers to the common systems that essentially
The rise of artificial intelligence (AI) has been talked about function as flow charts, but when most people are discuss-
in almost every industry this year, including finance. Large ing AI today, they're usually referring to the new genera-
language models (LLMs) like ChatGPT may even help open tion of generative AI pioneered by ChatGPT. These types
banking to bring it into the mainstream. It could lead to the of machine learning and decision-making algorithms are
"killer app" that will make open banking more accessible capable of creating "seemingly new, realistic content" (see
to the general public—just as MP3 players existed for years https://tinyurl.com/5984c27y). Therefore, you can ask LLMs
before the iPod became hugely successful and in turn to create 10 taglines for a new product launch and it can do
spurred the creation of smartphones. so, even producing entirely new content.

However, unleashing open banking’s full potential with For this article, I asked ChatGPT to write a paragraph about
the use of AI will still be a challenging process. the adoption of AI in open banking applications, and it pro-
duced the following:
How might open Banking evolve?
“The adoption of Artificial Intelligence (AI) in Open Bank-
Banks have been required to open their APIs to authorize ing applications has witnessed remarkable growth, revo-
third parties since the launch of the PSD2 directive in 2018, lutionizing the financial industry. According to a recent
and hundreds of companies have registered to use open study by McKinsey, AI applications in banking have the
banking protocols in the EU and UK. As of July 2022, 559 potential to create $250 billion to $350 billion in value an-
third-party providers were registered in Europe, and 221 nually. One of the key areas where AI is making an im-
of them were registered in the UK, reflecting the country’s pact is in fraud detection and prevention. A report by Ju-
strong standing in banking and technology (see https://ti- niper Research predicts that AI-powered systems will save
nyurl.com/kn7tn4rx). banks over $40 billion in annual fraud-related expenses by
2023. Moreover, AI is transforming customer experience in
Many of these registrations were either Account Informa- banking. A survey conducted by Accenture found that 79%
tion Service Providers or combined AISPs and Payment of customers are more likely to trust AI-powered recom-
Initiation Service Providers. Open banking may be prov- mendations for their banking needs. These statistics high-

18