The world of generative artificial intelligence (GenAI) can be a bewildering one at first glance. The closer you look at it, the weirder it can seem to get.
GenAI is more than just a complex technology evolution with serious potential to become ubiquitous in work and social life. It also has an esoteric philosophical battle happening behind the scenes. Proponents of GenAI argue that it will impact the ultimate fate of humanity; critics say it will primarily impact the fate of Silicon Valley.
Concepts such as artificial general intelligence (AGI), the technological singularity, effective altruism, the future of robotics, transhumanism, and so on make for great conversation topics (and subject matter for countless think pieces and blog posts). Yet, these topics are all secondary to the mundane day-to-day realities within banking that GenAI is well placed to transform. These more esoteric aspects of the philosophical underpinning of GenAI development are nonetheless fascinating. They will likely have a major influence on how this technology develops over the next few years. The recent drama at OpenAI and the ousting (and subsequent quick return) of CEO Sam Altman are one example.
At the center of this philosophical battle is a conflict between two schools of thought. The first is “effective altruism,” a modern rehashed form of utilitarianism that quickly goes to weird places, such as colonizing space with sentient artificial life and a strong concern over existential risks to humanity (e.g., rogue AGIs). Proponents here include Elon Musk and now disgraced and jailed Sam Bankman-Fried. The second is the “effective accelerationist” camp, sometimes abbreviated as e/ACC. Those on the e/ACC side argue that technology development and AI, in particular, are social equalizers. They believe the only moral stance is to develop these capabilities at top speed with minimal regulation so as not to impede progress. In the battle over OpenAI’s direction and management, the E/ACC proponents, backed by the significant funds and power of Microsoft, have essentially won out. Thus, the speed of GenAI development will only increase, barring any stringent regulations.
Much of the debate on GenAI focuses on these long-term and often morally grey areas, which risks distracting from the real-world impact these technologies can have in the here and now. Banking, in particular, is a key industry for GenAI, wherein routine and mundane manual processes that nonetheless remain technical and complex are major pain points for financial institution solution providers and their end users. Banking is, if anything, a practical industry. Outside of investors, it is less likely to focus on long-term projections of space colonization, AGI, or the technological singularity. Instead, banks want help with streamlining processes, generating efficiencies, and solving day-to-day problems. Areas such as payment exceptions handling, cash forecasting, reporting and analytics, and enterprise content management are the primary focus of banks today in deploying these capabilities.
GenAI providers who can help banks manage day-to-day pain points and ultimately build a clear business case will see significantly faster growth and deployment. If GenAI vendors do not focus on meeting these needs and solving actual business challenges, then many may dismiss GenAI as yet another technology hype cycle that they can ignore for the time being. Banks will need to focus on investigating and, in some cases, developing these practical use cases in conjunction with their technology partners, who will undoubtedly have less of an understanding of the needs of financial services.
Fortunately, the GenAI world is expanding rapidly, and the range of options and technology partners is advancing at a rapid pace. GenAI does not exclusively belong to OpenAI and Microsoft. Newer capabilities, such as small language models, and new players like France’s Mistral (an AI firm that achieved a double unicorn valuation of over US$2 billion on a Series A funding round), as well as development platforms like AWS Bedrock, and NTT Data, all grow the potential for banks to develop and launch GenAI solutions that can solve their practical problems.
As outlined in the recent Datos Insight Report Generative AI in Banking: Use Cases and Opportunities, less than a year after the public launch of ChatGPT, 60% of surveyed banks report they expect to launch GenAI capabilities within the next two years. That’s a shockingly high number for an industry typically seen as risk-averse and inherently conservative.
While the philosophical underpinnings of GenAI are fascinating (and frankly often alarming), banks and technology partners alike need to prioritize the day-to-day problems they face rather than get distracted by the esoteric hype and promises emanating from many.
