Transforming payments with generative AI technologies

Read the PDF

How GenAI is changing the payments game

Generative Artificial Intelligence (GenAI) is no longer just a futuristic concept. It is here, is becoming increasingly common, and is set to radically transform how we interact with technology in all areas. Payments is no exception.

GenAI is already reshaping the payments landscape by streamlining operations, enhancing fraud detection, and enabling hyper-personalized customer experiences. Looking ahead, it has the potential to transform how we handle money through its role in agentic commerce, where AI transacts autonomously on our behalf, acting as the decision-making engine behind autonomous digital agents.

In this deep dive into the next key influences in payments this year, the Visa Consulting & Analytics (VCA) team investigate GenAI, how it is shaking up payments, its economic impact, and what the future may hold for this revolutionary technology. Whether you’re a merchant, fintech, or a financial institution (FI), this is your guide to understanding GenAI in payments and how it may be relevant to your organization.

How GenAI is already transforming payments

As GenAI continues to evolve, its impact is being felt across every link of the payments value chain. Many companies, including FIs are experimenting with the technology to enhance their services, improve security, and offer more personalized experiences.

By mapping the payments value chain and applying a GenAI lens, we can identify the areas where use cases are being developed today and predict where disruption is likely to drive further innovation.

At leading organizations, GenAI has moved well beyond the planning stage; it is already being deployed across a variety of use cases, most of which generate returns through cost savings. In financial services, the initial focus has also been on internal operations, but new applications now include customer onboarding, fraud detection and prevention, and product recommendation and cross-selling.

Transaction Management

AI is deployed with real-time and data-rich context-based insights to drive excellent customer service and operational efficiency.

Examples: Real-time transaction analysis, Personal finance management, Triggers and insights

Fraud Risk Management

AI is commonly used to detect and prevent fraud by monitoring transaction activity and customer behavior to identify suspicious patterns and prevent loss.

Examples: Authorization and fraud detection and prevention, False positive fraud management, Scenario analysis

Customer Service

AI-powered virtual assistants and chatbots provide customer care, answer queries, translate FAQs, and even assess customer sentiment to optimize service.

Examples: Human-like GenAI chatbots, FAQ translators, Customer sentiment analysis

Disputes and Chargebacks

AI is increasingly being used to assist dispute agents and predict dispute and chargeback outcomes more efficiently.

Examples: GenAI to support dispute agents, Serial dispute prediction, Scheme rules support agent

Product Development

AI can innovate new banking products by analyzing current market trends, customer behaviors and competitor offerings to suggest product development strategies.

Examples: Customer Testing, Concept Design, Market Research

Product Management

Utilizing AI to better adjust product pricing to optimize consumer product appeal and ensure constant compliance with changing regulations.

Examples: Dynamic pricing, Real-time personalization, Data driven experimentation and analytics

Customer Sales

AI is currently being used to improve customer sales with improved lead generations and personalized product offers and cross-selling.

Examples: Personalized sales offers, Dynamic messaging and marketing, Product recommendations and cross-selling

Customer Onboarding

AI is currently being used to improve onboarding with hyper-personalized communication.

Examples: Pre-fill data automation, Hyper-personalized onboarding communications, KYC optimization

Customer Portfolio Management

AI can identify different customer segments within a bank's portfolio, and target them with personalized product recommendations based on behavioral predictions.

Examples: Activation, Usage, Retention

Transaction Flow

AI is used to achieve efficiency in account operations and processes.

Examples: Compliance automation, Documentation updates, Contract management automation

Some early and indicative use cases from across the payments ecosystem

Use case 1. Operational efficiency

Typically, FIs are adopting a low-risk approach to GenAI adoption, by starting with internal efficiency use cases and gradually moving to customer-facing applications.

Primarily, this is to reduce the risk of dealing with privacy concerns (which are more acute in markets, like Europe, where there is a strong focus on GDPR-like regulations), as well as reputational risks.

A typical application is automated document analysis, such as the automated generation of compliance documentation, policy updates, contract management, etc.

Itaú Unibanco, Brazil’s largest banking institution built a GenAI-enabled intelligent document processing solution, with a view to reducing operational friction, and lowering cost-to-serve. Incorporating 12 proprietary AI models, its tangible benefits have included faster onboarding and accelerated time to market.

Euroclear, the Belgium-based financial infrastructure group, has developed a 'legal co-pilot' that assists with regulatory research by turning legal questions into prompts and providing concise, natural language responses. It can be likened to a junior researcher that aids in navigating regulatory information, significantly reducing the time lawyers spend on research.

Use case 2. Customer onboarding

GenAI has significant potential to reduce onboarding times and costs by increasing the speed of identity verification and limiting the need for human intervention.

Airwallex, a Singapore-based payments and financial platform, uses a GenAI-enabled onboarding tool that reduced the occurrence of false positives by 50 percent and increased the number of customers that pass through onboarding without the need for human intervention by 20 percent.

Worldline, the France-based payment processor updated investors on its proprietary GenAI tool in 2024, saying that it would be accelerating its development. Among the use cases it highlighted was using GenAI to improve merchant onboarding.

Use case 3. Fraud detection and prevention

By analyzing transaction patterns in real time, GenAI can be far more adept than traditional rules-based techniques in spotting anomalies and flagging potentially fraudulent activity.

Visa has developed Visa Account Attack Intelligence (VAAI), a tool that uses generative AI components to identify and score enumeration attacks. The VAAI Score helps to reduce fraud and operational losses by assigning each transaction with a risk score in real time to detect and prevent enumeration attacks in card-not-present transactions. In the face of sophisticated application fraud risks, Visa has also enhanced its capabilities by integrating Featurespace’s real-time AI into its fraud prevention and risk-scoring offerings.

Use case 4. Customer experience - personalized offers, messaging, and campaigns

Building on traditional and predictive AI analysis of customer behavior, GenAI can formulate and offer tailored payment solutions, customized loyalty programs, and even personalized credit offers.

Klarna, the Sweden-based payments and buy-now-pay-later company, is reported to have run 30 marking campaigns during 2024 where GenAI was used to generate ideas, write copy and create images.

PayPal has developed an Advanced Offers platform to provide hyper-personalized shopping recommendations and discount offers.

Use case 5. Customer experience - virtual assistants

GenAI chatbots that support customer service or sales queries that mimic human-like interactions are becoming popular. These virtual assistants can handle customer inquiries 24/7, providing instant support and freeing up human agents for more complex tasks.

Commerzbank of Germany is one of the first European banks to have developed a virtual customer assistant, combining GenAI and avatar technologies in a customer application. Named Ava, the assistant enables the execution of transactions directly within the dialogue.

Use case 6. Regulatory Reporting and Compliance Automation

GenAI is being used to automate and enhance regulatory reporting, ensuring banks and payment providers remain compliant with evolving local and cross-border regulations. By parsing regulatory documents, extracting obligations, and generating tailored compliance reports, GenAI reduces manual errors, accelerates reporting cycles, and allows for proactive risk management.

Standard Chartered Bank of Singapore has partnered with Singapore-based RegTech firm Silent Eight to deploy AI-driven solutions that automate compliance investigations and regulatory reporting.

Use case 7. Financial Literacy and Customer Education

Banks and FinTechs in some regions are deploying GenAI-powered chatbots and content engines to provide tailored financial education and literacy programs. These solutions generate personalized learning modules, answer customer queries in local languages, and simulate financial scenarios.

Philippines-based bank RCBC has developed an app called DiskarTech that deploys AI-driven chatbots to deliver financial advice, explain banking concepts in Tagalog and other local dialects.

Use case 8. SME Credit Underwriting Using Alternative Data

In markets where small and medium-sized businesses (SMEs) often lack traditional credit histories, GenAI is used to analyze alternative data sources such as e-commerce sales, mobile wallet transactions, and utility payments for underwriting decisions.

Ant Group’s MYbank uses AI models to analyze real-time transaction data from Alibaba, Alipay, and other sources, enabling automated credit decisions for millions of Chinese and Southeast Asian SMEs, many of which have no formal credit history.

The potential economic impact of this transformation

From boosting revenue to reducing costs, GenAI could deliver significant economic benefits for payment providers and the broader financial ecosystem.

From a revenue perspective, GenAI is unlocking innovative business models and value-added services such as personalized financial advice. It is also enhancing customer satisfaction by improving automated customer interactions and helping guide human agents towards customized solutions in real time.

All in all, estimates suggest that, in terms of GDP impact, GenAI could add between $2.6tn and $4.4tn annually to the global economy.

What’s next?

The future of GenAI in payments looks bright, and its most disruptive impact is likely to come from its evolution into the engine of Agentic AI, autonomous digital agents that can decide, act and transact on a user’s behalf.

Selecting a sound strategic response

To unlock the full value of GenAI, there are several strategic approaches that payment players can pursue. These include:

1. Harness GenAI for operational efficiency and growth

Focus on utilizing GenAI to enhance operational efficiency, security, and customer experience, which could drive significant cost savings and revenue generation in the payments industry.

2. Develop a robust data strategy

To fully realize the benefits of GenAI, organizations need a comprehensive data strategy, one that is sponsored by senior leaders and aligned to business strategy.

3. Prioritize ethical AI practices

A growing emphasis on ethical AI necessitates that organizations protect customer data from fraud and cyberattacks.

4. Implement strong governance frameworks

Payment providers must adopt robust governance frameworks, conduct regular audits, and address ethical concerns to help ensure responsible use of GenAI.

5. Support ecosystem collaboration

Increased collaboration among payment providers, technology companies, regulators, and other stakeholders is essential.

6. Learn from the real-life experience of others

While GenAI has had a positive, transformative impact for many payment players, its implementation hasn’t always been a success.

7. Consider Build vs. Buy Approaches

Organizations should decide whether to build or buy GenAI.

8. Develop a GenAI-ready talent strategy

The success of GenAI initiatives depends heavily on having the right talent in place. Organizations should invest in building cross-functional teams that blend data science, ML/AI and prompt engineering, and GenAI-savvy product management.