# KYA Protocols Enhance AI Agent Reliability in Financial Transactions

> To combat the challenges of agentic AI in financial transactions, KYA frameworks are being tested to authenticate AI agents, ensuring trust and efficiency in automated workflows. Key players like Visa and J.P. Morgan are leading pilot projects to pave the way for secure AI-driven commerce.

**Source**: gfmag.com | **Published**: 2026-11-21 | **Type**: article

## Key Facts

- Visa and BILL's pilot reduces fraud alerts, enhancing AI agent transaction efficiency.
- Hyundai Card's AI agent manages liquidity, showcasing blockchain's role in treasury management.
- Know-your-agent protocols could lower false declines, improving financial transaction reliability.
- Agentic commerce growth hinges on balancing security and automation in payment systems.
- Collaboration among firms indicates a strategic shift towards AI integration in financial workflows.

## Summary

Recent initiatives by banks and payment firms to implement know-your-agent (KYA) rules signal a significant advancement in addressing the challenges of agentic AI adoption in financial transactions. This development is crucial as it aims to differentiate legitimate AI agents from fraudulent ones, a necessity for fostering trust in automated commerce. As the financial services sector increasingly integrates AI into workflows, ensuring the authenticity of these agents is paramount to mitigate the risk of fraud while promoting efficiency.

Current anti-fraud systems, which scrutinize transactions for unusual patterns, often misidentify legitimate AI activities as suspicious. Lev Brodsky, CTO of FinteqHub, highlights a common scenario: a transaction initiated by an AI agent at an odd hour may trigger alerts despite being routine for the agent's operational context. This paradox presents a barrier to the growth of agentic commerce, as existing fraud prevention technologies, designed to thwart automated attacks, inadvertently hinder the acceptance of AI-driven transactions.

To combat this issue, two pilot projects have emerged, focusing on embedding trust in AI workflows through KYA frameworks. One pilot involves a collaboration between Visa, BILL, and Billtrust, allowing AI agents to initiate approved invoice payments. In this setup, Visa's Trusted Agent Protocol authenticates the AI agent before the payment is processed. This initiative not only streamlines the payment process but also enhances security by ensuring that only verified agents can conduct transactions.

Simultaneously, J.P. Morgan Payment's blockchain unit, Kinexys, has partnered with Hyundai Card to explore treasury management through AI agents. In this pilot, an AI agent operates under a digitally signed mandate to manage liquidity in a blockchain deposit account. The agent's requests for foreign exchange transactions are subject to verification against predefined parameters, ensuring that only authorized actions are executed. This pilot illustrates the potential for AI agents to manage complex financial tasks while maintaining strict oversight.

While these pilots mark progress, they also highlight the limitations of current technology. As Brodsky notes, the next logical step for the industry is to integrate KYA protocols directly into payment systems and anti-fraud measures. Such integration would minimize false declines and enhance security, paving the way for broader acceptance of agentic payments. However, until this infrastructure is widely adopted, the scope of agentic commerce will remain constrained to scenarios requiring human oversight.

The implications of these developments extend beyond immediate operational efficiency. As financial institutions and technology providers refine their approaches to KYA protocols, the landscape for agentic commerce is poised for transformation. Companies that successfully navigate these challenges will gain a competitive edge, positioning themselves as leaders in the evolving market of automated financial transactions. 

In the coming years, the ability to seamlessly authenticate AI agents will likely become a critical differentiator for financial services firms. As trust in these systems grows, we may see a shift toward more autonomous AI-driven transactions, fundamentally altering the dynamics of commerce and financial management. This evolution will require not only technological advancements but also a cultural shift within organizations to embrace the potential of AI in decision-making processes.

## Entities

- **Companies**: FinteqHub, Visa, BILL, Billtrust, Kinexys, J.P. Morgan Payment, Hyundai Card, Hyundai Motor Group
- **Products**: Trusted Agent Protocol
- **Technologies**: AI agents, blockchain
- **People**: Lev Brodsky, Wee Kee Toh
- **Organizations**: Global Finance Magazine

## Key Concepts

agentic AI, know-your-agent rules, payment fraud, authentication, trust in workflows, pilot projects, liquidity management, anti-fraud systems

## Definitions

- **agentic AI**: AI systems that can autonomously perform tasks on behalf of users, particularly in financial transactions.
- **know-your-agent rules**: Frameworks designed to verify the identity and authority of AI agents conducting transactions.
- **Trusted Agent Protocol**: A protocol used to authenticate AI agents in payment processes.
- **pilot project**: A preliminary study or trial to test the feasibility of a concept or system before full-scale implementation.
- **liquidity management**: The process of managing cash flow to ensure that an organization can meet its short-term obligations.

## Use Cases

- AI agents initiating and completing invoice payments
- AI agents managing euro liquidity in a blockchain deposit account
- AI agents requesting foreign exchange executions

## Frequently Asked Questions

**What are know-your-agent rules?**

Know-your-agent rules are frameworks designed to authenticate AI agents in financial transactions. They help distinguish legitimate agents from malicious ones, ensuring secure transactions.

**How do pilot projects contribute to agentic commerce?**

Pilot projects test the implementation of know-your-agent frameworks, allowing AI agents to perform transactions securely. They provide valuable insights into the feasibility and effectiveness of these systems.

**What is the Trusted Agent Protocol?**

The Trusted Agent Protocol is a system used by payment processors like Visa to authenticate AI agents during transactions. It ensures that only verified agents can initiate payments.

**What challenges do financial firms face with agentic AI?**

Financial firms struggle to balance security with the need for efficient transactions. Modern anti-fraud systems can mistakenly flag legitimate AI agent activities as suspicious, hindering commerce.

**What is the next step for agentic payments?**

The next step involves embedding know-your-agent protocols into payment systems and anti-fraud infrastructure. This will help reduce false declines and enhance security for agentic transactions.

## Links

- [Read on Welcome.AI](https://welcome.ai/content/kya-protocols-enhance-ai-agent-reliability-in-financial-transactions)
- [Original source](https://gfmag.com/technology/know-your-agent-pilots-target-agentic-payment-fraud/)
- [Visa](https://welcome.ai/company/visa): Featured company

---

Source: Welcome.AI | https://welcome.ai/content/kya-protocols-enhance-ai-agent-reliability-in-financial-transactions