The subject of banks, treasurers and emerging technology continues to attract speculation and blue-sky thinking. But the signs are that we’re getting closer to viable solutions, and pilots and use cases are becoming thicker on the ground, as a white paper from BNY Mellon in early February testifies.
Entitled Multiple Paths, One Destination, the report on how artificial intelligence (AI) will benefit the relationship between banks and corporates cast a panoramic glance over the various technologies that are set to play a critical role in finance over the coming years, as banks and fintechs begin to collaborate with even greater enthusiasm and finesse.
In addition to real-time payments, SWIFT gpi, SWIFT Transaction Manager and distributed ledger technology (DLT), the report shone a spotlight on AI and its two most visible subcategories: machine learning and natural language processing.
AI, it noted, “is currently proving most effective in very specific use cases, including fraud monitoring, compliance and simple customer inquiries, paving the way for an AI-enhanced client experience and operational efficiencies”.
It added: “Banks are also exploring the potential to apply AI to higher-value activities – including treasury management functions such as liquidity management and payment channel optimisation – although such capabilities remain several years away.”
Not for the first time, AI is cited as something with great potential for payments – and yet somehow out of reach, waiting for the watershed of a ‘killer app’, or some similar breakthrough, before it can really make its mark.
In parallel, the institutional push for such a moment is growing stronger all the time.
In September last year, a McKinsey thought leadership piece urged banks to be “AI-first” – noting: “The potential for value creation is one of the largest across industries… AI can potentially unlock $1 trillion of incremental value for banks, annually.”
Disruptive AI technologies, McKinsey pointed out, “can dramatically improve banks’ ability to achieve four key outcomes: higher profits, at-scale personalisation, distinctive omnichannel experiences and rapid innovation cycles.”
With that in mind, it warned: “Banks that fail to make AI central to their core strategy and operations – what we refer to as becoming ‘AI-first’ – will risk being overtaken by competition and deserted by their customers.”
The piece recalled a 2018 white paper from fintech player Finastra, in partnership with Accenture, entitled Payments and API Banking: Riding the Third Wave of API Innovation to Enable the Digital Economy.
From that vantage point, Finastra forecast “a new era of ‘platformification’ in banking and payments services”, driven by open APIs in the context of cloud computing and AI.
That new generation of services, it noted, would be “driven and enabled by an open, collaborative ecosystem of innovative API providers [and] characterised by a remorseless focus on delivering the optimal payments experience for customers”.
Those who believe that this constant parade of promise is simply white papers talking to themselves should know that the era these thought leaders have predicted is gradually dawning. The hype mother ship is starting to land.
Here are three, recent innovative partnerships and ventures that are making tomorrow happen today:
A subsidiary of Europe’s largest utilities business the Enel Group, Enel X Financial Services is an electronic money institution, authorised by the Bank of Italy to provide payments solutions to consumers and corporates.
In September last year, the firm announced that it had selected AI-driven fintech player Tink as its open banking technology provider.
Through a single API, Tink enables clients to access aggregated financial data, initiate payments, enrich transactions and build finance management tools. It is now devising technology solutions for a range of Enel X customers in Italy and Europe.
Enel X Financial Services CEO Giulio Carone said: “Through the partnership with Tink we will be able to support our clients in the daily management of their finances, with an innovative and engaging solution that uses machine learning to provide tailored and personalised advice.”
Unveiled in October, this platform uses AI to organise data and sharpen its quality – with the aim of slashing weeks- or even months-long reconciliation processes to mere seconds.
An upgrade to the original SmartStream Air launched in 2018, the new version features a built-in ‘observational learning’ system powered by AI software called Affinity.
This enables the platform to understand how different payment records correlate with one another, as well as to learn – and even mimic – the actions that users take with the records they oversee.
As a result of those capabilities, Affinity significantly reduces the time it takes to match complex data sets.
Importantly, according to SmartStream chief innovation officer Andreas Burner, working with the user interface requires no specialist IT training or expertise. “You don’t even need to understand the data,” he says. “Affinity knows how to compare complex data sets and the results are achieved in seconds.”
The platform is positioned as a dependable ally in any effort to weed out discrepancies from particularly large batches of payment records.
Some readers may recall that in our news roundup of November last year, we covered the expansion of a partnership between financial institution Commerce Bank and treasury tech provider HighRadius.
Aimed squarely at corporates, the fruit of the partners’ beefed-up collaboration has been branded Integrated Receivables.
As Commerce Bank recently noted, AI plays a “big part” in the service, enabling corporate accounts receivable (AR) staff to work more efficiently and achieve a higher automatic on-invoice hit rate.
For example, it explains, the decoupling of remittance data during the electronic payments process is a major resource drain – particularly as AR is often tasked with manually matching payments to remittance data. Not only is the process arduous – it is highly susceptible to human error.
Integrated Receivables remedies multiple AR challenges, making it possible to:
The platform’s AI provides end-to-end automation by capturing remittance data from multiple sources, such as email attachments, paper invoices, customers’ payment web portals and other types of electronic payment files. To reduce future payment exceptions, the AI also learns payer behaviours.
Platform by platform, the future is arriving.
Matt Packer is a freelance business, finance and leadership journalist