The future of artificial intelligence (AI) in banking is brighter than ever, as the adoption of AI-based solutions continues to gain pace. The potential for value creation for the financial services sector is immense, as AI could unlock an estimated $1 trillion of incremental value for banks annually, according to calculations by McKinsey & Co.
While predictive AI has already made some inroads in such areas as fraud detection and risk assessment, the full potential of Generative AI is yet to enjoy broad adoption across the financial services sector. I believe that innovative banks, particularly digital-first players like TBC Uzbekistan, are at the forefront of big changes in the industry as they work to overcome the two greatest roadblocks that have slowed Gen AI’s widespread adoption in banking. By the end of next year, the obstacles outlined below will likely significantly diminish, ushering in a new era of innovation and efficiency across banking.
Go deeper with GlobalData
The cost of Generative AI adoption
Banks are already widely applying predictive AI to risk scoring, fraud detection and Next Best Offer (NBO) models, which leverage data-driven insights to tailor product recommendations to individual customer needs and preferences. However, it’s been harder for financial institutions to benefit from the latest breakthroughs in Generative AI, given that most of the solutions are cloud-based, and banks have not been able to use them because of the restrictions and concerns related to sharing sensitive personal data externally.
Luckily, some institutions with the right engineering talent, including TBC, have been able to build their own AI infrastructure and use open-source pre-trained models to incorporate Gen AI into their products. This is a real step change, precipitated by falling cost of innovation, that is proving hugely important because it allows relatively smaller tech-enabled players to unlock the potential of Gen AI technology for their specific business needs.
Banks have a wealth of great data to leverage, and the more forward-looking players are focusing on building their own architecture that uses open-source pre-trained LLMs. The multi-modal capabilities of these LLM-based solutions – text, audio, text-to-speech and speech-to-text – have the power to disrupt and improve conventional customer communications channels. Smaller players and large financial institutions working in areas where so-called ‘low-resource languages’ are prevalent don’t have to stay in the margins and can also take advantage of the latest breakthroughs, thanks to declining costs.
Unlike languages such as English or French, these ‘low-resource languages’ often lack quality Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) models, which are prerequisites for using Large Language Models (LLMs) in voice channels. A prime example is Uzbek, the language spoken in Central Asia’s most populous nation. To overcome this at TBC, where we operate the region’s largest digital bank, we used our own data to build our proprietary speech tech models (which is now relatively cheaper to do) and launched them to further increase our efficiency and improve the customer experience.
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Thank you!
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form
By GlobalDataWhile banks still face some hurdles in deploying and operating all the necessary components of Gen AI tech on proprietary infrastructure, the positive changes in the relative affordability of adopting the latest tech innovation is ushering in a new ‘’post-mobile’’ wave of fintech disruption. Building AI infrastructure necessary for training and inference of models for proprietary speech tech is an expense that virtually any bank can now afford to undertake, and I believe that this level of investment is likely to be returned relatively quickly via greater operational efficiency.
Trustworthiness of AI
Another big challenge banks face in AI adoption is the issue of trustworthiness, specifically preventing hallucinations—errors where AI generates false or misleading information. In a customer-facing environment, such mistakes can damage trust and lead to potentially severe compliance issues, particularly in highly regulated industries like banking.
Many experts in the field, including companies like Aiphoria.ai, Verax AI and our own team at TBC, are working on solutions for making generative AI results trustworthy for large-scale deployment by businesses. While the problem has not been fully solved, this work is showing impressive results with AI systems making great strides in reducing hallucinations through rigorous training on proprietary data and refining model architectures.
The future is now
To overcome the remaining hurdles and help harness the full potential of AI, banks globally are already allocating 22% of their budgets on average towards AI, with spending on AI set to rise by 6.3% according to the Infosys Bank Index.
Advancements in AI are allowing banks and other fintechs to embed the technology across their entire value chain. For example, TBC is leveraging AI to make 42% of all payment reminder calls to customers with loans that are up to 30 days or less overdue and is getting ready to launch other AI-enabled solutions. Customers normally cannot differentiate the AI calls powered by our tech from calls by humans, even as the AI calls are ten times more efficient for TBC’s bottom line, compared with human operator calls. Klarna rolled out an AI assistant, which handled 2.3 million conversations in its first month of operation, which accounts for two-thirds of Klarna’s customer service chats or the workload of 700 full-time agents, the company estimated. Deutsche Bank leverages generative AI for software creation and managing adverse media, while the European neobank Bunq applies it to detect fraud.
Even smaller regional players, provided they have the right tech talent in place, will soon be able to deploy Gen AI at scale and incorporate the latest innovations into their operations. Next year is set to be a watershed year when this step change will create a clear division in the banking sector between AI-enabled champions and other players that will soon start lagging behind. We will witness a transformation of AI from a tool for increasing efficiency to one that is a critical driver of innovation in banking. For example, AI will likely help banks launch new products that address the pain points of their customers much more precisely than human-generated ideas, enabling them to bring these new products to market faster. I feel privileged to be part of this truly revolutionary moment in our industry.
Konstantin Kruglov is Head of AI at TBC Bank Group
Related Company Profiles
McKinsey & Co Inc
Infosys Ltd
Klarna Bank AB
Deutsche Bank AG