By Vilas Madan, Senior VP and Growth Leader of APAC, EXL
Artificial intelligence (AI) is slowly revolutionising almost every industry, and banking is no exception. AI offers a host of opportunities and tools that can be used to deliver faster services, more personalised customer experiences, and boost operational efficiency at a scale weโve never seen before. But for all its promise, many banks are still struggling to tap into AIโs full potential. The barriers they face are significantโand overcoming them is essential if banks want to keep pace. From my perspective, there are four key challenges holding Australian banks back.
1. Integrating Data and Intelligence
AI is only as good as the data that powers it. Banks today handle huge amounts of data, but all too often, this data is siloed across departments or tied up in legacy systems. Bringing together structured and unstructured data into a unified, high-quality dataset is no small task, yet itโs essential for AI to deliver on advanced use cases.
Without proper data integration, banks cannot leverage AIโs full potential. As highlighted in a recent EXL-supported study by Everest Group, effective data and intelligence integration can unlock a staggering 41% revenue growth. But to get there, banks need a strong data foundationโone that connects internal and external data sources with security, accuracy, and accessibility.
2. Addressing Talent Shortages in AI Expertise
The talent required to implement and scale AI is in short supply. Australian banks need specialists such as data engineers, prompt engineers, AI architects, and domain experts, but the competition for these roles is fierce, which poses a significant barrier to AI adoption. This shortage means banks must find creative workarounds, whether upskilling their existing workforce or forming strategic external partnerships.
According to Matt Coates, technology lead for Accenture in Australia and New Zealand, โOnly a third of executives have a clear vision for how generative AI will impact their workforceโ. This means that if leadership wants to fully realise the power of reinvention offered by genAI, they must adopt a strategic approach that โharmonises technology and human talentโ. Collaborating with experienced providers like EXL can help banks bridge the talent gap, leveraging pre-built solutions and proven expertise to accelerate AI adoption.
3. Viewing AI as a Core Operating Model, Not Just a Tool
AI shouldnโt be seen as standalone technology or a simple series of experiments. It needs to be deeply embedded into a bankโs core operations. When viewed this way, AI can become a transformative operating model that drives scalable success.
The key to unlocking sustainable value from AI lies in blending different models into a cohesive system. For instance, instead of deploying AI only in isolated processes, banks can use it to orchestrate end-to-end workflows. This approach, known as AI orchestration, creates lasting value by improving efficiency and reducing costs. Banks that treat AI as a strategic enabler rather than a technical add-on will achieve the most meaningful results.
One such use case is customer onboarding, where generative AI can dramatically improve efficiency in customer onboarding workflows. From automating records aggregation to reducing redundant information requests, data-driven AI models can make integration challenges less daunting while simultaneously improving customer satisfaction.
4. Overcoming Cultural Resistance
One of the biggest challenges in AI adoption isnโt technical; itโs cultural. Introducing AI is a big change that often requires significant shifts in organisational mindset and operations. Banks must align stakeholders, train teams, and ensure the organisation is ready for change.
Itโs equally important to maintain human oversight, particularly in areas like compliance, risk management, and customer service. AI is powerful, but it canโt fully replace human judgment. Keeping humans in the loop ensures accountability, fosters trust, and minimises ethical risks, a principle we at EXL prioritise in every implementation.
A Way Forward for Australian Banks
To succeed, banks must adopt a pragmatic, phased approach to implement AI solutions. In doing this, Australian banks can lead the charge in AI innovation while reaping huge benefits. For example, at EXL, our generative AI solutions have delivered significant results, from processing 30 million underwriting documents with 90-97% accuracy to reducing call-handling times by nearly 50%.
As outlined in EXLโs “5 Keys to Fast-Tracking Generative AI,” starting with clear strategies, preparing strong data foundations, and incorporating human oversight are critical steps. Itโs best to start with small, measurable projects and scale gradually to build momentum while managing risks. For more insights and support, visit EXL at www.exlservice.com.
