Over the past year, many of the largest players in the FinTech industry have introduced solutions that are in one way or another connected to artificial intelligence (AI) agents. Autonomous AI agents are of particular interest because they don’t just assist users, they replace them. Such systems gain agency, they don’t merely recommend or execute specific instructions but independently assess situations and make decisions.
Let’s try to understand whether the current hype around AI agents is just a short-term trend or if what’s happening is a systemic transformation of the FinTech industry.
AI agents are software systems that use artificial intelligence algorithms to perform tasks such as data analysis, natural language processing, or decision making. Typically, these programs operate according to predefined instructions, functioning as intelligent assistants.
Autonomous AI agents, on the other hand, can independently formulate goals, plan actions, and adapt to changes, allowing them to perform more complex tasks without constant human intervention.

Examples of AI Agent Implementation by FinTech Companies
In 2024–2025, many major FinTech players, including payment networks, trading platforms, payment service providers, digital asset issuers, neobanks, software providers, and others, introduced AI agent-based solutions or the infrastructure to work with them. Here are several specific cases:
- Mastercard launched the Mastercard Agent Pay platform, which allows verified AI agents to make tokenized payments on behalf of users.
- Visa introduced the Visa Intelligent Commerce platform, which opens access to the company’s global payment network for AI agents.
- Tether announced the launch of an open-source AI platform for autonomous agents capable of making payments using USDT and BTC in a P2P network without APIs or centralized infrastructure.
- Coinbase developed the open standard x402, enabling AI agents and applications to make payments in stablecoins directly via the HTTP protocol.
- Stripe announced the Smart Disputes tool, which uses AI agents to autonomously handle chargeback claims.
- Revolut announced the deployment of AI agents to initiate financial actions based on client behavior, as well as automate user request handling, limit management, and subscriptions.
- PayPal released the PayPal Agent Toolkit developer kit, which enables the creation of autonomous AI agents capable of securely performing commercial transactions via the PayPal API.
- Block introduced a framework for developing autonomous AI agents with open source code, among others.
Moreover, many leading international banks are also adopting similar solutions. For example, Mitsubishi UFJ Financial Group (MUFG), in partnership with Sakana AI, is developing a platform of autonomous AI agents that can independently create complex documents, including internal reports, regulatory documentation, and more, as well as conduct comprehensive research.
Another example is the NextGen Online Assistant (NOA) AI agent based on the Amelia cognitive platform, implemented by BNP Paribas on the NeoLink client platform. The system independently processes operational tasks in real time, including inquiries about settlements, fund movements, market information, and more.
These cases clearly demonstrate that AI agents already represent a significant component of product and infrastructure solutions in the FinTech industry. But what has driven the market to this shift right now?

Key Drivers Behind Demand for Autonomous AI Agents in Finance
A major shift in the development of neural networks and related technologies occurred over a decade ago with the advent of machine learning based on big data.
All current AI technologies are technically classified as narrow or weak artificial intelligence and operate based on neural networks. In essence, all popular AI-powered chatbots, such as ChatGPT, Gemini, Perplexity, and others, are user interfaces for interacting with neural networks.
Around 2022–2023, the field of AI technologies experienced a significant leap forward, specifically:
- A new generation of large language models (LLM) emerged: GPT-4, Claude, LaMDA, and others. These became more reliable, faster, and more accessible, enabling the development of real products based on them.
- Open-source frameworks for working with LLMs were introduced: LangChain, AutoGen, CrewAI, and others. This made it easier for even small teams to develop multi-agent and tool-integrated systems.
Large language models (LLM) are AI algorithms trained on massive volumes of text data, capable of “understanding,” generating, and processing natural language.
During this same period, the financial sector saw an explosive increase in both the volume and complexity of operations. Operating 24/7/365 and constantly expanding use-case scenarios became the standard for FinTech systems. At the same time, competition and customer acquisition costs were rising, while profit margins were shrinking. This situation pushed financial companies to seek new ways to boost operational efficiency, making AI technologies a timely solution.
Autonomous agents help bridge the gap between the intensity of business processes and the operational limitations imposed by the human factor. They enable real-time automation of routine tasks such as user support and verification, transaction routing, and commission optimization. AI agents also reduce risks in decision-making chains, especially in complex B2B scenarios involving compliance.
On the user side, people increasingly expect to simplify their interactions — to set a task and forget about it. For example, a user might say: “Get me travel insurance for a week in Turkey,” and the agent will independently select the product, handle payment, and provide all the necessary documentation in the proper format. This is what the UX of a modern financial app looks like when aligned with current market expectations.
All of this is unfolding amid the development of new regulatory frameworks and growing interest and trust from institutional investors and major corporations. For instance, Google has unveiled a new online shopping mode where an AI agent tracks product prices, adds items to the cart, and can complete purchases on behalf of the user via Google Pay. The solution enables a fully automated shopping experience, including virtual try-ons and personalized preferences.
The conclusion is clear: the implementation of AI agents in financial solutions isn’t just a passing trend but a systemic transformation — one that will only continue to gain momentum in the coming years.