Small bank, big moves: How a Maine-based bank is bringing over a thousand employees on its AI journey

Similarly, organizations with robust data governance principles in place will already have the oversight, accountability, policies, quality improvement methods, and understanding of organizational data assets that can be applied to GenAI use cases. “Applying GenAI to market analysis can reliably support and supplement human analysts,” says Lees. “This accelerates their work while detecting trends, and delivering, potentially, more accurate market predictions.” That leadership could prove temporary, however, and may be just as likely to benefit traditional banks. LLMs can exhibit unpredictable behaviors, especially when exposed to novel inputs. This unpredictability can pose risks in compliance scenarios where consistent and reliable outputs are essential.

By automating and accelerating the extraction of key information from customers’ loan application documents, loan officers are now able to make faster, more accurate and informed decisions in approving loans. KPMG in the US developed and deployed a customized Gen AI solution, using language models (LLMs) trained within the bank’s secure environment to ensure the highest standards of confidentiality of sensitive customer information. The artificial intelligence revolution is well underway, but how ready are banks and lenders to leverage the full breadth of these capabilities? Fintech sits at the crux of change, powering a shift from Gen AI hype to Gen AI implemetion, as banks first see the value of having great data, and then look to onboard to right tools to access and use it in ways they could never have before.

This platform prioritises security, reliability and robustness to support the widespread adoption of generative AI across CaixaBank’s operations. Tom Hewson, CEO at RedCompass Labs, said, “From a process, skill, and expertise point of view, payments innovation and market share are being seized by a few big banking players who are widening an already significant competitive gap. The strategic investment Bangor Savings Bank has made in this initiative is indicative of how aware it is of the tides of change. Using Gen AI doesn’t have to be a harbinger for job cuts – instead, it can be the cause for banks to adapt to a new role in the community as a place for learning skills that improve its people’s employability. The recent paradigm shift brought about by Gen AI has reopened many debates about de-skilling and job insecurity. Surveys that report 54% of roles in banking are at risk of job displacement don’t help either.

A bank employee wanting to understand how a certain type of customer might respond to a proposed offer first creates a target persona, such as a 20-to-30-year-old female professional living in a large city. The Artefact solution uses the target persona to model a virtual “cluster” of customers, with each cluster representing 2 to 3 million real customers. The employee can then interact conversationally with customer avatars generated by IBM watsonx.ai AI studio, querying them about their personal preferences and consumption habits. The insurer teamed up with IBM Business Partner® TUATARA to reimagine its customer service experience. In one month, Generali Poland rolled out Leon, a virtual assistant built with action.bot from TUATARA, based on IBM watsonx Assistant.

Registration closes at 5 pm central time Tuesday, November 5

Karim Haji, Global Head of Financial Services, outlines why it’s such an exciting time for the financial services industry. What’s better, however, is when you can integrate genAI across a broader process. Apply genAI across the process and you can start to run the various steps in parallel. And these kinds of applications could deliver productivity gains of, say, 75 percent.

In the future, these co-pilots could tailor investment strategies in real-time or predict market trends, helping to fortify FS firms’ competitive edge and deliver differentiated client outcomes. Temenos Generative AI solutions in core banking operations enhances workflows and day-to-day queries, allowing banks to innovate in product creation and account management. It empowers business users to interact with data using free text speech, providing insights in a simplified manner. This promises to significantly reduce the time spent on such tasks, enabling banks to concentrate on optimising operations and improving customer experiences.

Chances are, the last time you dealt with your financial institution, artificial intelligence was already involved. You may have had a question answered by a digital assistant, or received a personalized marketing offer, or even been the beneficiary of rapid market analysis. The fact is, tomorrow’s financial service winners and losers may be determined, in large part, by how effectively they’re able to deploy and scale GenAI applications today. “AI algorithms analyze vast amounts of data to assess credit risk, detect anomalies, and prevent AML fraud,” Saxena notes. Earlier this year, the Paris-based Financial Action Task Force removed the UAE from its “grey list” for deficiencies in money laundering controls, a move that drew criticism from some anti-money laundering analysts.

Impact summary and future directions

In banking, these risks can manifest themselves in areas ranging from an inaccurate basis for decision-making to discrimination against particular population groups. These lapses can not only cause severe reputational damage, but also lost opportunity costs. In particular, boards may be reluctant to use GenAI-generated analysis in their decision-making or sign-off AI use cases because they don’t have sufficient confidence in the outputs.

He also suggested banks “pre-position” collateral to ensure central bank support should something go wrong. At the beginning of June 2024, we launched Cora+, the first generative AI pilot which introduces that new technology into Cora for the first time. We hope to be able to answer a larger number of customer questions more succinctly and make fewer unnecessary hand-offs to colleagues. The first offered basic help and support which was instructional – guiding customers on how to complete tasks. The second generation in 2020 was more personalised, recognising the customer at an individual level and allowing them to undertake transactions like reordering a PIN, changing their address or reactivating an ISA, for example.

Governance complexities with RAG implementations

Generative AI supports IT development by automating coding tasks, generating code snippets, and assisting in quality assurance processes. Additionally, AI plays a crucial role in modernizing legacy systems, enabling them to support advanced applications and meet evolving business needs. LLMs play a crucial role in risk management by analyzing transaction patterns, identifying suspicious activities, and generating alerts for potential compliance violations. This enhances the institution’s ability to detect and respond to financial crimes swiftly. Global financial institutions must navigate a complex landscape of data privacy regulations, ensuring that their AI systems comply with varying requirements across jurisdictions. This involves implementing robust data governance frameworks, ensuring data anonymization and encryption, and maintaining transparency in data processing practices.

You could be forgiven for thinking that suddenly everyone seems to be talking about GenAI. In its original form, AI was about machines learning and improving iteratively themselves – not necessarily being fully trained by people. Generative AI uses a different type of technology to deliver different outcomes. Deutsche Bank has collaborated with Kodex AI, a Berlin based start-up company that  focuses on developing Artificial Intelligence (AI) solutions, to launch the whitepaper  ‘Adopting Generative AI in Banking’.

But banks clearly understand the urgency; a huge majority are already dedicating resources to GenAI. For smaller and midsize organizations in earlier stages of GenAI adoption, a CoE will suffice as a first step and coordination point for knowledge. Further, a CoE will allow the organization to incrementally improve capabilities, spread best practices, foster knowledge sharing and promote early use cases.

The question now is what will financial services do next and how soon will they apply AI across the entirety of their organizations and more broadly with customers. And then the third force that is helping drive down cost is captured by Wright’s Law, which says that as we work on and use a new product, our improved skills and processes reduce costs by a fixed proportion each year. gen ai in banking Major banks like JPMorganChase, which recently launched a suite of generative AI tools to help more than 50,000 of its employees work more efficiently, are likely to gain an early advantage from this move. Mastercard’s insights come at a key time too, with banks expected to significantly ramp up their integration of Gen AI into back and front-end services over the next year.

It has the potential to lift the banking industry out of its technological doom loop and enable leading adopters to establish deeper and more meaningful relationships with their customers. Where it gets amazing is when it starts to fundamentally change ‘the possible’. It’s where the productivity gains get to a point where you can start to do things you never thought possible.

And by supporting customers with more of their transactional queries, Cora is freeing up time for colleagues to have quality conversations with customers in the moments when they really need that care, empathy and consideration. Our recent global research survey gives insights into key strategies and applications for GenAI in banking, and we look forward to sharing the results at Sibos 2024, the annual conference and exhibition organized by Swift. Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited (“DTTL”), its global network of member firms, and their related entities (collectively, the “Deloitte organization”). DTTL (also referred to as “Deloitte Global”) and each of its member firms and related entities are legally separate and independent entities, which cannot obligate or bind each other in respect of third parties.

This process helps ensure the accuracy and reliability of AI-generated content and decisions. These include documentation requirements for AI models, evaluation methods for AI systems, and implementation controls. Gone are the days of chatbots that appeared on a website and provided standard answers to specific questions from customers, without actually improving the service but rather irritating the customer. This website is using a security service to protect itself from online attacks.

We have also set up a responsible AI taskforce comprising senior leaders from multiple disciplines to assess and address these risks prior to any use case being deployed in production. Our existing responsible data use framework for AI continues to provide us with guardrails as we look at new use cases. This framework, called Pure, ensures that our use of data remains purposeful, unsurprising, respectful and explainable for customers. In addition, we are building a technology infrastructure to enable the adoption of large language models in a secure manner. At the same time, it was necessary for the bank to adhere to a countless number of regulatory requirements.

Of course, with any new technology comes challenges, so Mastercard outlines how banks can mitigate these new obstacles. Discover how AI revolutionizes consumer experiences and boosts business efficiency in India. Biggest AI concerns – Their five biggest concerns are user expertise (29%), low-quality inputs/outputs (28%), security and data protection (27%), transparency of decision-making (25%) and accuracy of AI algorithms (25%).

As with any new technology, the advent of GenAI brings about a natural sense of curiosity and adjustment for our employees. We are determined to bring our employees along by focusing on the synergy between human and AI capabilities to leverage GenAI as a co-pilot. In institutional banking, we tapped GenAI to help reduce the time needed for relationship managers to fill in the ESG (environmental, social, and governance) risk questionnaire by summarising company reports and prepopulating relevant fields. It has also paved the way for further applications of Gen AI throughout the bank’s operations, and the potential for even greater gains in productivity and operational excellence. Starting with cost, potential users of the technology stand to benefit greatly from a combinatorial effect caused by three powerful forces.

For financial services, AI needs to not only pass the litmus test at an institutional level, but it must also have buy-in from customers. Customers are only going to be comfortable with AI being at the heart of their financial planning tools or lending decisions if there is a clear benefit. Partners that understand the organizational pain points, as well as the distinct needs of a bank or lender’s customers, will drive smoother implementations.

This capability is useful for pairing customer caches with historical trend data to inform risk assessments or flag anomalous transactions indicative of potential fraud. The bank’s customers are benefitting as well from a substantial reduction in the time required to process their loan applications, from days to under one hour. This improvement is not only enhancing customer satisfaction by speeding up loan approvals, but also reducing the bank’s operational costs and increasing its capacity to handle a larger volume of loan applications.

Looking ahead, CaixaBank plans to analyse and develop new use cases as part of GalaxIA. Identifying a use case necessitates substantial effort in prioritization, cost-benefit analysis, and strategic considerations regarding technology and data architecture. Therefore, financial institutions worldwide are typically exploring only 7-10 crucial use cases on average.

Banks, nonbank financial services players, and FinTech or Big Tech are at different stages of the journey to harnessing Gen AI capabilities. Established financial institutions are experimenting with Gen AI use cases, initially in marketing and sales, customer support, risk and compliance. You can foun additiona information about ai customer service and artificial intelligence and NLP. They have improved product ChatGPT search and client service capabilities and have initiated change programs to overcome the obstacles posed by data quality issues, fragmented processes and systems, and legacy risk policy frameworks. Beyond customer service, generative AI in banking is also transforming fraud detection and risk management.

• In the near term, banks should focus on driving forward the highest value potential opportunities while factoring in the level of risk exposure.
• Another 38% said they plan to incorporate it into their business within the next 24 months.
• Organizations must consider when and how employees can leverage GenAI and evaluate the distinct risks of internal and external use cases.
• Anti-Money Laundering (AML) and Global Financial Compliance (GFC) frameworks are foundational to maintaining the integrity of the financial system.

Bahrain and Dubai are positioning themselves as Islamic finance hubs, and applying generative AI would seem a natural progression that could have global implications for the two tech-centered economies. With fintech valuations still high, the likelihood of traditional banks acquiring their upstart rivals is questionable. And venture capital, the main source of funding for many fintechs, is also under pressure. Data from PitchBook shows a downward shift in investor sentiment that might slow further funding rounds.

The BIS predicts all of its members will adopt generative AI to enhance their internal cybersecurity measures. Central banks that have already implemented generative AI have praised its effectiveness in detecting cyber threats compared to traditional tools. The BIS believes in the potential for widespread adoption of GenAI, an area in which many central banks have developed a strong interest. We are at a historic cusp in time, and ChatGPT App we will all need to navigate how GenAI figures in the way we live and work. On DBS’s end, we have in place a governance structure that will help us balance reaping the benefits of Gen AI while managing the risks of a still-emerging field. Temonos’ new suite of Gen AI solutions is not just significant for its banking clients, it also points to a wider industry trend where Gen AI is being actioned rather than merely spoken about.