Q&A with Michelle Bendschneider, Industry General Manager of Financial Services (A/NZ), DXC Technology
Q1. How do you see AI reshaping the fintech industry over the next 12–24 months?
A: AI is moving out of the lab and into the very core of fintech operations. We’re seeing a shift from experimentation to real, measurable impact. The biggest change will be the compression of decision cycles, which will lower costs and accelerate time to market. This forces fintechs to rethink how work gets done, how products evolve, and how risk is managed. Fintechs that adapt quickly will thrive, while those stuck in pilot mode will fall behind. The next 24 months will be about operationalising AI at scale, not just testing its potential – and to do this it will require a longer-term strategy for change.
Q2. Many fintechs have experimented with AI. What changes in 2026 as adoption scales across the business?
In 2026, you’ll start to see AI embedded in the day‑to‑day: underwriting, compliance, customer experiences, payments, servicing, financial crime operations, and more. We will see a shift from siloed experiments and isolated wins to reusable platforms, governance, integrated capabilities across functions and scalable frameworks. There will also be the emergence of agent managers, and human-in-the-loop processes start to become part of the operating model.
Q3. Why are compressed decision cycles becoming such a defining pressure for fintech leaders?
Because markets aren’t slowing down. Customer expectations, competition, and risk dynamics all demand real‑time responses, and AI adoption becomes a competitive disadvantage and advantage. Human‑only processes simply can’t keep pace at scale. Fintechs that are adopting AI are able to move faster without compromising control or quality. This sets a new standard and speeds things up across the board. I like to compare this to how we all went from navigating with paper maps to using real-time GPS maps for smarter, safer and faster navigation, and now find it hard to turn back.
Importantly, AI doesn’t do well with the historic cadence of decision making. Waiting three to six months to completely analyse a use case is almost antithetical to AI. By the time a paper exercise is done the use could have been stood up, validated and scaled; not to mention the technology (and teams) will have moved on.
Q4. What is agentic AI, and why does it represent a step change rather than incremental automation for fintechs?
Agentic AI is an advanced form of AI that is capable of autonomous decision making and goal-driven action with minimal human supervision. It perceives context and can proactively solve complex, multiple problems through observation, reasoning and learning. So it has the ability to redesign how fintechs operate end-to-end in a major structural shift, connecting data, systems and workflows. You’ll see step-change gains in speed, cost, and customer outcomes, not just marginal efficiency.
Q5. Where do you expect agentic AI to have the earliest and biggest impact within fintech operations?
To my earlier point – if you focus on the impact on workflows, I’d say we’d see the most immediate value in high-volume, decision-heavy areas, and three places stand out. The first is in risk management and fraud detection, where real-time pattern recognition reduces false positives and losses. AI agents can adapt and respond dynamically, analyse risk profiles, flag anomalies and trigger interventions in real time. The second would be payments, where AI agents can optimise routing and resolve disputes, improving conversion and recovery and reducing friction across payment flows. Finally, customer servicing. AI agents can resolve routine inquiries end-to-end, freeing human teams for higher-touch cases.
Q6. Why do so many fintechs struggle to move AI from pilot stages to real, enterprise-wide impact?
There’s often a disconnect between ambition and infrastructure. While the modius operandi (MO) is to start small, the folly is often that many pilots aren’t designed to scale – no legacy integration, limited data access, minimal governance. Also, governance and operating models haven’t kept pace with technical innovation. Enterprise impact needs production‑grade data pipelines, clear ownership and controls, and a services model so AI can be deployed repeatedly across use cases. The MO should really be “start small, and solve for scale”
Q7. How important is legacy system integration in making AI work at scale for fintechs?
It’s critical. Most core systems won’t be replaced overnight, so AI must work with existing infrastructure. That means orchestration, APIs, and consistent workflow design. When integration is first‑class, AI outcomes become predictable, auditable, and easier to scale. Interestingly, with this in mind and the sprawl of “Stove-pipe AI” (AI that sits on top of single sources of data or workflow), we’re seeing a shift in focus to Agent2Agent (A2A) and Model Context Protocols (MCPs). These approaches provide open standard to connect agents, enabling horizontal AI to solve for federated applications and data.
Q8. With so much AI hype in the market, what will truly differentiate fintech leaders in 2026?
Execution and a long-term plan. The hype will fade, but the winners are those that demonstrate reliable, explainable AI in production, meeting regulatory expectations and improving unit economics. The fintechs that prove AI can work securely, responsibly and repeatedly, while aligning AI investment to clear business outcomes, will pull ahead. Rather than hundreds of disparate POCs, a longer-term plan is required to strategically compound value. Partnerships are also critical to success and fintechs who work strategically with SaaS, Cloud and SI partners have the most to gain when it comes to achieving solutions at scale.
Q9. Why are governance, explainability and operational execution becoming non-negotiables?
AI is a risk but it’s also a solution to a much broader set of risks. Governance is absolutely table stakes because trust is the currency fintechs trade in. Partners, customers and regulators need to understand how decisions are made and how risks are controlled. They all want confidence that AI is being used responsibly, and poor governance is both a compliance issue and a business risk. Explainability, robust monitoring, and model lifecycle management will turn AI from a promising idea into a dependable system, accelerating approvals and partnerships in the process.
Q10. What advice would you give fintech leaders today who want to be competitive in 2026?
It is a strategic effort, not a tactical technology experiment… Start scaling now with clear use cases in mind that compound, and mobilise as if it’s a strategic imperative (e.g., the board needs to own it, not the technology team). Focus on the outcomes that matter most. Focus on the use cases that deliver clear value and build the minimum set of capabilities to deliver them repeatedly: clean data paths, integration, and governance. And most importantly, treat AI as a core operating capability and embed it into the fabric of how your business runs.
