Unzer on the future of AI in payments and its impact on PSP operations

Niv Liran, Chief Technology Officer at Unzer, discusses how AI is redefining the payments industry and driving greater organisational efficiency.

AI is a hot topic across the payments industry. From Unzer’s perspective, how will it shape the future of ecommerce, and what benefits could it bring to merchants and shoppers?

AI is going to change ecommerce in a big way. What we’re seeing now is the rise of what we call ‘agentic commerce.’ This means that AI agents will soon handle end-to-end commerce, from search and discovery all the way to item selection and checkout, with limited or even no user input.

For consumers, that means a level of convenience we’ve never seen before. Imagine telling an AI what you want for your holiday trip, and it goes off to research your past bookings and preferences, finds the best deal, books it, and pays for it, all without you lifting a finger. That saves time and reduces abandoned checkouts.

For merchants, the impact is just as significant across all aspects of the business. Conversion rates go up because hesitation and checkout friction are reduced. Businesses can also use AI to manage supply chains, inventory, and even product photography (by using GenAI), as well as listing and pricing automatically. The big change is that merchants won’t just focus on human shoppers anymore; they’ll need to make their products easily accessible to AI agents that can find and buy automatically and ensure they work with a payment provider that enables a native and transparent AI-based checkout experience.

Our focus is on building secure infrastructure and helping merchants adapt to AI-driven commerce, while digitising their processes and keeping them competitive in this new landscape.

Beyond payments, in what business areas is AI at Unzer being applied?

We’ve been experimenting with AI across many parts of the company. Our developers use it to write and test code. The legal department uses it to compare contracts and highlight areas of interest and differences to our policies, something that used to take hours can now be done in minutes. Compliance teams use AI to check regulatory requirements against our guidelines, which helps us stay aligned with regulators. Customer service also benefits, as AI suggests answers to customer requests based on past cases, helping new colleagues get up to speed much faster and enabling experienced ones to focus on more complex topics.

AI also levels the playing field. For example, I’m not a lawyer, but with AI I can read and understand legal language, compare contracts, and even spot points for renegotiation. That’s why I strongly believe AI is here to support and empower people.

And what lessons can be learned from these concrete applications?

AI isn’t a magic bullet. It works best when it tackles repetitive, time-consuming tasks where there is a good organisational knowledge base to train on. It doesn’t replace human expertise; it frees employees from routine and simple work so they can focus on more complex problems that need judgment, creativity, and experience.

What cultural and organisational changes are needed for successfully adopting AI?

I think the biggest change is cultural rather than technical. At Unzer, we realised early on that adoption depends on having compelling use cases. Our approach has been to make AI accessible to everyone. We call this ‘democratising AI’. By giving lawyers, compliance staff, sales teams, and others the ability to build and use their own AI applications, we increase adoption and get better-fitting AI use cases. We’re also making it easier to build agents with simple no-code or low-code tools.

Additionally, we introduced AI ambassadors in each department. These are early adopters who train their colleagues, show them practical use cases, and bring back new ideas. For example, I’m in tech and know a lot about AI models, but I wouldn’t know which processes a compliance employee finds cumbersome and relevant for AI assistance or automation. That’s where these ambassadors help bridge the gap.

A concrete example is our Quarterly Business Review (QBR) agent. Preparing a QBR for a customer used to take hours of gathering data and creating slides. Now, the AI does it in minutes, and the employee only needs to review and fine-tune it. That saves several working days each month.

Training the teams is also key. Employees need to see that AI is there to support them. Once people experience how much time they save on repetitive tasks, they start to see AI as an opportunity to focus on more meaningful work. The cultural shift is really about embedding AI into daily workflows and making employees feel confident using it.

How should companies balance trust, risk, and responsibility when using AI, particularly in highly regulated environments like the payments industry?

Trust is critical, especially in fintech. All stakeholders, be it merchants, their end customers, or we as Unzer, need to know that their data is safe and that access is tightly controlled. That’s why we focus on building security and trust from the ground up.

First, the need-to-know principle is key. Our AI models work with permissions based on the user and the use case, and no single AI agent can access all of the company’s information. For example, the legal agent can only access legal information, but not transaction or merchant data.

Second, transparency is essential. If our legal team asks AI about a contract clause, the system doesn’t just give an answer – it also shows the source so employees can check the original text. The same applies to questions about compliance or other topics.

Third, compliance is non-negotiable. As a regulated financial institution, we follow frameworks like the EU AI Act closely. We avoid high-risk applications and focus on enabling people to work faster and smarter, safely.

Looking ahead, the payments industry will need new standards for AI agents, like how they’re authenticated, how security tokens are handled, and how fraud is prevented in a 24/7 AI-driven environment. Until then, the responsibility is on providers like us to ensure security and protect consumers.

What is your outlook on the future of AI in commerce and the payments industry?

Looking ahead, I see AI moving from assistive tools to AI that can act on its own. Agentic commerce will become mainstream. Payments might happen automatically, marketing will increasingly focus on AI agents instead of humans, and the whole shopping experience will change.

On the merchant side, we will see more back-office business processes, such as inventory management, merchandising, product photography, and listing being increasingly executed by AI with human supervision. 

The key factor will be trust. People will only let AI make decisions if they feel confident it acts in their best interests. That’s why transparent, responsible, and compliant AI systems will set successful companies apart. For Unzer, it’s both a challenge and an opportunity to create a payments ecosystem where AI doesn’t just automate tasks but really improves the experience for merchants and consumers alike.

About Niv Liran

 

Niv Liran is the Chief Product and Technology Officer at Unzer, where he leads the company’s product and engineering teams and shapes its technology strategy. Under his leadership, Unzer has built a unified-commerce platform that brings together online and in-store sales, giving businesses a complete, real-time view of their customers and operations. He also guides Unzer’s work in artificial intelligence, overseeing the company’s AI roadmap and initiatives. 

 

About Unzer

Unzer helps companies go digital with simple and integrated payment and software solutions. Whether it’s shopping in-store, on mobile, online, or handling payments and daily business tasks, Unzer offers everything businesses need in one place – an ecosystem that makes retail simpler, more efficient, and seamless for consumers. More than 85,000 merchants across Europe already use Unzer's ecosystem. The company employs around 750 people in eight offices in Germany, Austria, Denmark, and Luxembourg.