ARTICLE
Banking on intelligence: the AI opportunities for 2026 and beyond
Generative AI has moved from curiosity to core infrastructure in less than three years. Since the launch of ChatGPT in November 2022, banks and financial institutions have accelerated adoption of large language models across customer experience, fraud detection, software engineering and data operations. What began as controlled experimentation is now scaling enterprise-wide – and the next phase is already emerging.
For banks, the commercial opportunities are vast. AI is turbocharging processes that, even in the age of digital banking, remain cumbersome and error prone. But as three senior executives from Publicis Sapient, Santander UK and Deutsche Bank emphasise, the real prize lies not just in automation, but in reimagining how banks operate, compete and grow.
AI moves from automation to acceleration
AI-powered automation is not new in banking. What has changed is the sophistication, speed and breadth of what modern models can handle – and the growing confidence of banks to deploy them.
The productivity gains are compelling, says Jan-Willem Weggemans, VP, Executive Client Partner of Publicis Sapient: ‘Tasks that once took months are now being completed in days or even hours’.
But the impact of AI isn’t just measured in efficiency metrics. Standardisation is improving quality and customer confidence not by making banking robotic, but by freeing people to focus on higher-value work requiring judgement, expertise and empathy.
Customer service teams – long burdened with manually interpreting emails, complaints and documentation – are increasingly supported by AI. Modern systems convert unstructured inputs into workflow-ready data, enabling real-time product recommendations, hyper-personalised engagement and proactive next-best-action alerts. The downstream impact is significant: smoother operations, faster turnaround times and more consistent outcomes.
And it’s not just the customer-facing operations that are turbo-charged by AI. At Deutsche Bank, copilots and internally developed tools are helping engineers accelerate everyday work. As a result, the bottleneck is moving from writing code to knowing how to instruct AI to produce it effectively.
‘We’re seeing that impact starting to ramp as we train and educate more people,’ says Tim Mason, Managing Director, Artificial Intelligence at Deutsche Bank, where AI projects fall under three internal pillars: ‘augment, automate and amplify’.
The rise - and reality - of agentic AI
Banks may still be absorbing and adapting to generative AI but already a new wave of disruption is incoming: agentic AI. Widely seen as the next frontier, agentic AI refers to systems that can autonomously execute multi-step tasks, collaborate across workflows and deliver end-to-end outcomes. In banking, this could transform complex journeys such as loan origination or customer onboarding.
Jasmeet Narang, Chief Transformation, Data and AI Officer at Santander UK, points to the loan process as a prime example. Historically, multiple teams handled credit checks, fraud screening and compliance reviews, with friction at each handoff. An agentic system could gather documents, run affordability checks, trigger fraud screening, request missing information from the customer and prepare a conditional approval – all with minimal human intervention.
However, the technology remains early-stage and comes with health warnings. ‘You have to be mindful that it requires a different type of mindset, the right controls and careful monitoring of the outputs of these agents,’ says Narang. ‘There’s a lot of experimentation and proof of concept right now. I’m not sure anybody is doing it at scale with multiple complex processes.’
Jan-Willem Weggemans at Publicis Sapient agrees. ‘Everyone is talking about it but not so much doing it,’ he notes. His advice: start with a “good enough” minimum viable version, put it into production and learn from real-world performance.
Accuracy thresholds will vary by use case. A travel recommendation agent might tolerate 90% accuracy; a payments agent cannot. Banking’s inherently high bar for precision when it comes to managing a customer’s money means progress will rightly be measured and incremental. According to Weggemans, many banks may opt to work hand-in-hand with trusted partners to get this right and piggy back on existing work that has already derisked the new technology. He highlights Sapient Bodhi, his company’s enterprise-grade agentic AI platform, which uses a library of pre-built, compliant-by-design agents¹.
‘The pre-built agents are fine-tuned by capability, such as hyper-personalised customer engagement or mortgage back office orchestrator, so banks can put them to work straight away or customize for their unique needs,’ he says. ‘Bodhi accelerates their transition from pilots to mass production.’
New revenue models on the horizon
AI’s longer-term impacts could be far more disruptive. One of the biggest constraints in banking today is the cost and scale of personalised customer interaction. AI changes that equation. Hyper-personalised, always-on engagement could unlock new product models and deepen customer relationships.
Wealth management is a particularly promising area. More intelligent guidance, dynamic portfolio insights and real-time risk assessment could encourage customers to interact more frequently and invest more actively. Over time, this may reshape customer behaviour – and the bank’s revenue mix.
Tim Mason of Deutsche Bank sees AI as a partner in unlocking new markets. Instead of hiring entire teams to enter areas like new bond instruments, AI can spot opportunities, cut costs, streamline work, and help focus on driving revenue
‘It changes the dynamic of what you can do,’ says Jan-Willem Weggemans of Publicis Sapient. Jasmeet Narang of Santander agrees: ‘We could see completely new business models emerge,’ he says.
Governance by design
With greater use of AI comes increased awareness of both its potential and its risks. Yet the challenge of ‘responsible AI’ grows as the technology grows in scope, sophistication and power. Already there’s been a subtle shift – if recent years were all about keeping ‘a human in the loop’, then, in 2026, we expect to hear a lot more about ‘the human over the loop’ as scaling creates capacity constraints.
In this model, with plenty of well-understood controls and guard rails, AI handles the heavy lifting while humans retain ultimate accountability and review. For example, an AI system may compile a KYC assessment, but a human still makes the final onboarding decision.
At scale, however, banks will also need AI systems to self-report performance metrics to satisfy model risk management requirements, such as the Federal Reserve SR 11-7 requirements in the US, which requires comprehensive documentation across the model lifecycle, independent validation and regular reviews to detect drift or degradation.
‘These rules may be challenging because manual validation every six months won’t be feasible at scale,’ says Tim Mason of Deutsche Bank. ‘Automated tools and solutions will be the only way to fulfil this regulation. It’s about being safe without being slow.’
Banks are keen to work with regulators so that rules are robust without acting as a brake on innovation. Santander UK, for example, is already working closely with regulators and is part of the Financial Conduct Authority’s AI Live Testing initiative launched in December 2025.
The message from industry leaders is clear: proactive regulatory engagement is now essential.
Augmenting the human workforce
Despite persistent fears about job losses, early advocates who framed AI as augmenting workers rather than replacing them are being vindicated by the use cases now in deployment. Investment research, for example, gets the amplifier effect when background analysis is automated, allowing teams to produce more timely and relevant reports. Instead of reducing headcount, many institutions are instead expanding coverage and generating more relevant and timely content for clients. Analysts can monitor more companies, publish insights faster and respond more dynamically to market developments.
Jobs may not disappear wholesale as a result of increased deployment but they will change. Banks have a responsibility to ensure the workforce has the education and skills to make the most of the opportunities – those already opening up, and those not yet even imagined. Measuring usage intensity – such as daily interaction levels – provides early signals of where adoption is working and where additional support is needed.
As employees gain in confidence and use the new tools to solve real problems in their day-to-day work, resistance to the technology eases – and productivity blooms.
This skills uplift starts at the top, with leading bankers agreeing that AI literacy is rapidly becoming a core leadership requirement. ‘You can’t be a successful leader unless you have at least an intermediate understanding of AI and what it means for your organisation,’ says Narang of Santander UK, which has embedded this philosophy into its leadership development through an AI Talent Booster programme, including reverse mentoring for its top 250 executives. The goal is to build confidence, curiosity and practical understanding.
Tim Mason of Deutsche Bank agrees, pointing out that executives do not need to become data scientists, but they must understand enough to identify opportunities, challenge assumptions and lead informed debates. ‘It’s a rare capability but highly trainable,’ he adds.
Leaders must also manage the human side of transformation. As AI capabilities mature, banks may need to rethink organisational structures and workforce composition.
Jan-Willem Weggemans foresees flatter organisations with fewer layers. In technology teams, AI-supported developers could shift the talent mix toward more senior profiles capable of orchestrating intelligent tools rather than performing routine coding. The same dynamic may apply to management. With AI providing deeper, real-time insight into performance, leaders can spend less time investigating and more time acting. Interactive, drill-down reporting powered by AI could fundamentally change how organisations are steered.
However, the transition must be handled carefully. Workforce impacts are real, and proactive reskilling pathways will be essential to maintain trust and engagement.
From experimentation to reinvention
The industry has largely moved past the question of whether AI will matter. The focus now is on how quickly – and how responsibly – banks can scale it to deliver enterprise value.
The winners are likely to be those that balance ambition with discipline: investing in agentic capabilities while building robust governance that doesn’t slow innovation and pursuing efficiency while keeping sight of new revenue opportunities. AI may have started as a tool for automation. Increasingly, it is becoming a catalyst for reinvention – both for people, and the banks that employ them.
Want to know more? Jan-Willem Weggemans, VP, Executive Client Partner, Publicis Sapient, is a keynote speaker on Day 1 of MoneyLive, where he will discuss ‘Rewiring the Bank: the path to enterprise scale’.
Don’t miss this opportunity to hear practical insights on how to effectively and compliantly scale in the real world.
MoneyLIVE, 9-10th March, Business Design Centre, London
Jasmeet Narang
Chief Transformation, Data and AI Officer
Santander UK
Tim Mason
Managing Director, Artificial Intelligence
Deutsche Bank
Jan-Willem Weggemans
VP, Executive Client Partner
Publicis Sapient
MoneyLIVE is your one-stop shop for all the latest banking and financial services content from across the world. With a huge portfolio of conferences, webinars, executive roundtables and reports to choose from, MoneyLIVE provides our community with relevant and future-facing information.
Visit the website: https://moneylive-insights.com/
Publicis Sapient is a technology company that provides enterprise AI platforms and services. With over 30 years of digital business transformation experience, we enable enterprise clients to transform how they operate and serve their customers, unlocking new value and enabling them to thrive in an AI-driven world. Our platforms use AI built off this deep enterprise context to help them modernize, build agentic solutions, and sustain their competitive advantage. The combination of our AI platforms and the expertise of our people enables us to deliver faster and more effective outcomes through solutions that are specific to the unique needs of our clients’ businesses, their industries and their customers. Publicis Sapient is the technology hub of Publicis Groupe, uniting 20,000 people worldwide. For more information, visit publicissapient.com.
Stay in the loop with MoneyLIVE
Subscribe to our newsletter to receive news, insights and special offers.