How is AI Affecting the Banking Industry?

You don’t have to be an avid reader of business news to have come across ChatGPT. The new open-AI start-up has been capturing headlines everywhere, featured in credits to poems that circulate in social media and sparking discussions in politics about how to handle seemingly omnipotent autonomous AI tools. It is no wonder then that the conversationalist artificial intelligence programs have re-ignited the question: what can AI do for banks? 

Composing text is only one facet of AI and it is probably the least interesting one for banking. Sure, it is easy to imagine where tools like ChatGPT could boost efficiency in banking services. It could be deployed to make chatbots more helpful while sounding more human or voice assistants better understand queries. It could support employees with writing compliance procedures or creating executive summaries. ChatGPT, for example, is an AI sub-discipline called generative AI that produces content, whether that is in text, images or music. Yet the value of AI in finance is much broader than that and what banks have been looking at for decades are quantitative applications. 

Application fields for machine learning and predictive analytics 

Some of AI’s core concepts have been around since the 1950s, but it is only in recent years that we have seen a revived interest in it. Exploding computing capacity, driven by a constantly increasing performance of semiconductors, met exploding data pools. This combination meant that machine learning models became ever more potent and predictive analytics so powerful that they could be used across the entire spectrum of banking operations. Here are some of the most promising application fields: 

  • Customer segmentation and cluster generation 
  • Predictive customer behaviour 
  • Risk analysis 
  • Fraud detection 
  • Investment algorithms and algorithmic trading
  • Robo-advisory 
  • Language and emotion processing 
  • Process automation 
  • And many more 

The huge range of application fields tells you how game-changing AI can be. Just like blockchain and distributed ledgers, AI is a so-called general-purpose technology. Comparable to the invention of electricity or the steam engine, these are technological breakthroughs that will impact the entire economy on a fundamental level, not just one or two industries. Banking is at the core of both of these tectonic shifts. In case you are interested how the two are working together in banking, I have written about it here 

Banks were early adopters of AI and machine learning because theirs is a quantitative discipline that was able to crunch data even before computers could make sense of texts or images. This is why those techniques are already being deployed efficiently, say when analysing risk probabilities or stock performances. This is nothing to be taken for granted. Many other AI applications, though promising, cannot deliver the same performance as current methods. This is due to a phenomenon called the productivity paradox, which basically describes how it might take decades until a new technology becomes more efficient than the old one. But AI is not just a boon to banks. Like every game-changer, it opens the gate for other players to handle (parts of) the banking business more efficiently. 

Big Tech betting big on artificial intelligence 

The five technology titans – Alphabet, Meta, Apple, Amazon and Microsoft – have been working hard to leverage technological advances to break into the financial services industry. While the blockchain is a natural fit for their efforts that gravitate around finance, research on AI has been broader. Still, it also encompasses some application fields where banks are using AI. 

Google used to be the master of AI. Its CEO Sundar Pichai is well-known for his AI-first mantra. Today, all Big Five are putting it at the top of their agendas. The Economist quantified this commitment on multiple levels: in 2022 – a tough year in terms of tech-stock valuations – they pumped a combined $223bn into research and development (vs. $109bn in 2019), of which a large chunk flowed into the various forms of AI. That is more than a quarter of their combined sales. The Big Five are also pouring money into AI through acquisitions and investments, of which around a fifth involve AI companies. About 10% of all job postings require some AI skills. And the investments are accelerating. In 2022, on average, there was one AI investment per month, which is three times as much as in the preceding years. Not all tech titans are equal in their commitment to AI, though. According to the analysis, Alphabet and Microsoft are leading the roster, closely followed by Meta.   

Unlike blockchain and cryptocurrency investments, much of the money that flows into AI is not directly related to Big Tech’s goal of entering financial services. The aforementioned ChatGPT, for example, is a start-up partly owned by Microsoft. Alphabet is launching a similar rival called Bard. Amazon spends much of its AI budget on its profitable cloud computing business. Meta and Alphabet focus heavily on using it to improve their ad business. Voice assistants like Apple’s Siri or Amazon’s Alexa are also benefitting from tech titans’ AI dollars.  

Probably the most prominent example of AI for finance is Apple’s 2022 acquisition of the British open-banking fintech Credit Kudos, an expert in forecasting the likelihood that credit applicants will be able to repay their loans. It was quickly turned into a market-ready functionality and in 2023 the Colossus from Cupertino launched its Apple Pay Later offering in which it used Credit Kudos AI to do its own risk analysis, thus venturing deep into traditional banking turf. 

The one thing banks must get right 

Just like with every other technological advance, big or small, there are two sides to the coin when it comes to the rise and proliferation of artificial intelligence. Act quickly and focus on the areas that matter and you will be able to capture more market share and boost your profitability. Act too slow or back the wrong horse and competitors will outpace you faster than you think. There is no simple solution to how to make the most of the AI wave. Whether banks make the right decision in timing, focus and investment amount depends on a host of (often individual) factors, but one thing is certain: they shouldn’t be waiting to see what the others will do, because then it will most certainly be too late to get into the AI game.