This is the third part of the multi-part blog series by AICorespot, on emergent, innovative technologies that are in pole position to drive the financial and banking space in the very near future. This part of the blog goes into detail on issues such as implementation, and fitment.

Banks and financial enterprises are currently spoilt for choice, due to the sheer number of emergent and trending technologies available at their disposal. While the saying ‘more the merrier’, does hold good in tech options in some ways, a broader selection of choice puts enterprises in a bind as they struggle to evaluate for fitment. Analysis, paralysis, if you will. Fitment can be defined as fitment to the environment, company philosophy, and fitment to the existing tech stack. When choosing a trending emergent technology, enterprises need to ask themselves the following three important questions.

1) Is my purchase a good fit for the environment?

2) Will it integrate well in the company’s current tech stack?

3) Does the tech align with company philosophy? Is it anti-ethical to the company’s values ?

Answering these questions helps banking and financial enterprises hone in on specific products, therefore facilitating efficiency in the process. But when navigating the sea of options available at their disposal, companies should be wary of not falling into the hype trap, all technologies should be solutions-oriented, hype is often empty noise or usually innovation that fails to be solution-oriented. Making the wrong investment in technologies like this is something that banking and financial enterprises ought to avoid.

The current market scenario with regards to emergent technologies in banking and financial enterprises presents an unparalleled opportunity. Historically, the court of public opinion had deemed that all-in-one solutions were the ideal, but with the advent of emergent technologies such as IoT, a more holistic opinion has begun to prevail. We previously spoke about right fitment, but there is another layer that is very important in the decision-making process, present-day tech is developed from the ground up to be compatible and in general, as open platforms. Most critically, they are developed to be part of an ecosystem, and to be complementary to other technologies.

What banking and financial enterprises are discovering is that the right combination of technology paves the way for radical transformation and innovation. AI has revolutionized the industry as being a sort of “constellation of innovation” – broadly encompassing technologies such as NLP (of particular interest to the banking space), Computer Vision, ML, and Deep Learning. At the heart of it all, of course, is the data. Sherlock Holmes, the famous Victorian detective, once famously quipped “Data, data, data, I cannot make bricks without clay!”, and never has that saying been more relevant that in the current day.

What these technologies do is that they help banking and financial enterprises harness the data they have available at their disposal to the fullest extent. For example, ML in Fraud Detection has made some serious inroads and has demonstrated great promise, and the tech in itself is old to some companies, and new to some companies.

Specifically speaking, NLP and other such language technologies have witnessed explosive growth during the time of the pandemic, as in-person visits to banks have drastically reduced. Chatbots have picked up in a massive way, and while they have significant advantages in that they help in replacing a human agent with a more efficient, cheaper, AI-based agent, they do have their caveats.

It’s not all sunshine and roses, however. Research demonstrates that humans will always be ahead of AI in terms of activities that require our rich human imagination. In that same vein, what AI-/ML-based chatbots struggle with is understanding the implications of human context and nuance. This is where the crippling fault with AI lies, they are utterly tone deaf with regards to understanding human context and nuance. This is one of the reasons why AI technologies in the context of chatbots can be viewed of as being hype.

AI-based drivers in organizations have historically required to be supported by SMEs. Presently, it is being broadly deployed in a very off-the-shelf-way, which applications making extensive utilization of AI/ML technologies. The customer is closer to these technologies than ever before, and influencing their experiences in tangible, positive ways is the immediate objective. In the transfer of funds for instance, AI/ML is harnessed to improved user experiences by customizing the UI in accordance with the behaviour and moods of the client.

Why banks ought to become AI first

Over the previous few decades, banks have, on an ongoing basis, dynamically adapted to the advances and tech innovations to redefine how clients interact and communicate with them. Banks put forth the idea of ATMs into action in the 1960s, and electronic card-driven payment mechanisms in the 1970s. The turn of the new millennium witnessed widespread proliferation and adoption of 24/7×365 online banking, which was quickly succeeded by the app-based, mobile-driven “banking on the go” paradigm in the 2010s and 2020s. However, with the second decade of “fingertip banking”, AI is poised to revolutionize the segment by shifting paradigms once more. Service- and company-orientation are a thing of the past, a customer-facing orientation is the essence of present day software, and technology in general.

It is general consensus that we’re currently in an era of AI-driven digitization, enabled by reducing expenditure in relation to data storage and process, universal access and connectivity, and swift evolution in AI technologies. These emergent technologies can herald an era of increased automation, and during deployment after adjusting for risks, can usually enhance on decisions made by human agents both in terms of quickness and precision. The prospect with regards to value creation is one of the biggest opportunities across industries, as could possibly unlock close to $1 trillion of incremental value as far as banks are concerned, year-upon-year.

On a range of more than 25 use cases, “AI tech, and being in position as AI first can prove to be a positive force in assisting the boosting of revenues through enhanced customization and personalized experiences being served to clients (and staff members alike), reduced expenditure, and more efficient and streamlined resource utilization, and unravel new and prior untapped avenues on the basis of an enhanced capacity to process and produce insights from massive reserves of data.

In a broader sense, disruptive AI technologies can drastically enhance a bank’s capacity to accomplish four critical outcomes, increased bottom line, at-scale customization and personalization, unique omnichannel experiences and swift innovation cycles. Banks that do not step up to the plate and make AI fundamental to their core strategies and operations – what is currently being thought of as being “AI first”-will be taking the risk of being overtaken by contestants and abandoned by the clientele. This risk is further underscored by four prevailing trends:

• Increasing client expectations as adoption of digital banking increases.

As the COVID-19 pandemic dawned upon civilization, utilization of internet-based and mobile, application-based banking channels, across the globe, has increased by an approximate 20% to 50% and is predicted to sustain at this relatively high level once we’re past the pandemic. Throughout divergent global markets, ranging between 15% and 45% of clients predict that they will be cutting back on in-branch, in-person visits, following the conclusion of the COVID situation.

As clients increase their engagement with digital and electronic, app-based banking services, they become more demanding, specifically in contrast with the standards that are the norm from leading consumer-internet businesses. Meanwhile, these market influencers raise the bar on customization and personalization on an ongoing basis, to the extent where they at times predict client requirements even before the client themselves can identify them, and provide highly customized, personalized services at the correct time, through the correct channels.

• Leading financial enterprises leverage sophisticated AI tech is gradually increasing.

Approximately 60% of financial-services domain survey respondents in McKinsey’s Global AI Survey reported that their businesses had embedded at the minimum, one Artificial Intelligence functionality/capacity. The most typically leveraged AI tech are: Vas (Virtual Assistants) or Conversational Interfaces (32%) for client service divisions; and ML strategies (25%) to identify fraud and support underwriting and risk administration. While for several financial services enterprises, the harnessing of AI is episodic in nature and concentrated on particular, targeted use cases, an increasing number of banking forerunners are taking a comprehensive stance towards the deployment of advanced AI, and integrating it across the complete lifecycle, from the front-to the back-office (Exhibit 2)

• Digital Ecosystems are disintermediating conventional financial services

Through facilitating access to a divergent grouping of services through a typical access point, digital ecosystems have revolutionized the way clients discover, assess and purchase goods and services. For instance, WeChat users within China can harness the same app not just to exchange messages, but additionally to book a cab, get food, schedule a spa appointment, engage in recreation, send money to a contact, and access a private line of credit.

Likewise, across the planet, nonbanking enterprises and “super apps” are integrating financial services and products within their journeys, providing compelling and innovative experiences for clients, and disrupting conventional strategies for discovering banking products and services. As an outcome, banks will be required to rethink how they take part within digital ecosystem, and utilize Artificial Intelligence to leverage the capacities of data to its fullest extent available from these new sources.

• Tech giants are getting into financial services as the subsequent adjacency to their core business models.

Internationally, leading tech giants have developed significant market advantages; a large and engaged client network, massive reserves of data, facilitating a robust and increasingly accurate comprehension of individual clients, organic strengths in generating and scaling innovative technologies (which include Artificial Intelligence); and access to low-cost capital. Historically, tech heavyweights have aggressively made an entry into adjacent businesses looking for fresh revenue streams of offerings. Heavyweight players have already obtained a foothold within financial services in chosen domains (particularly within payments, and in a few scenarios, insurance and lending), and they might soon seek to press their benefits to deepen their presence and develop massive scale.

Hurdles to AI implementation within the banking sector

Banks are currently encountering two groups of objectives, which seemingly are at odds. On one side of things, banks are required to accomplish the quickness, agility, and flexibility native to fintechs. This is being done with initiatives such as AI first. On the other side, the ought to continue managing the sale, security conventions, and regulatory requirements of a conventional financial-services enterprise.

Regardless of billion of dollars expenditure on charge-the-bank tech initiatives every year, very little banks have succeeded in diffusion and scaling of AI tech across the enterprise. Amongst the hurdles hindering bank’s efforts, the most typical is the absence of an obvious strategy with regards to Artificial Intelligence. Two extra hurdles for several banks are, to start with, a weak core tech and data backbone and, secondly, an outmoded operational model and talent strategy.

Developed for stability, bank’s core tech systems have featured good performance, specifically in assisting conventional payments and lending ops. But, banks must identify a resolution to various drawbacks native to legacy systems prior to them deploying AI tech at scale (Exhibit 5). To start with, and foremost, these systems usually don’t have the capability and flexibility needed to support the variable computational requirements, data-processing requirements, and real-time analysis that closed-loop AI applications need.

Core systems are tough to change, and their upkeep needs considerable resources. What is more, several banks data reserves are spread across several silos (independent business and tech teams), and analytics efforts are concentrated narrowly on stand-alone utilization scenarios. With no centralized data backbone, it is virtually not feasible to undertake analysis of the appropriate data and produce a smart recommendation or offer at the correct moment. If data makes up the bank’s basic raw material, the information ought to be governed and made available securely in a manner, that facilitates analysis of data from internal and external sources at scale for millions of clients, in almost real time, at the “point of decision” throughout the enterprise.

Finally, for several analytics and advanced AI-models to scale, enterprises require a solid group of utilities and standardized procedures to develop, test, deploy, and monitor models, in a repeatable and “industrial” ways.

Bank’s conventional operational models further impede their attempts to meet the requirement for ongoing innovation. Most conventional banks are organized around distinct business lines, with centralized tech and analytics teams built as cost centres. Enterprise owners define objectives unilaterally, and alignment with the business’s tech and analytics strategy (where it exists) is usually weak or inadequate. Siloed operational teams and “waterfall” implementation procedures invariably cause delays, expenses overruns, and suboptimal levels of performance.

On top of this, enterprises are lacking a test-and-learn mindset and solid feedback loops that put forth rapid experimentation and iterative enhancement. Usually not content with the performance of historical projects and experimentation, business executives have a tendency to be reliant on third-party tech providers for crucial functionalities, starving capacities and talent that ought to ideally be generated in-house to ensure market differentiation.

Conclusion

This brings us to the conclusion of the third part of this blog series on Emergent tech that are set to drive the financial and banking fields in the immediate future. The next part in this blog series will talk about off-the-shelf banking applications that integrate AI/ML, RPA and how its deployed in banking/financial services. We will providing a nuanced perspective on RPA implementation and deployment within the banking and financial services domain.