AiNxt and AtOM are designed to bring AI-driven automation to enterprise workflows, payment-system integration, partner onboarding, testing and certification across UPI and other digital payment platforms
NPCI Steps Up AI Push
The National Payments Corporation of India (NPCI) has launched two new agentic artificial intelligence (AI) platforms — AiNxt and Agentic Orchestration & Messaging (AtOM) — as it looks to bring greater automation and intelligence to India's rapidly expanding digital payments ecosystem.
The platforms were unveiled at the Global Fintech Fest 2026 in Mumbai and are designed to automate enterprise processes that currently involve significant manual intervention.
The launch marks another step in NPCI's broader effort to incorporate emerging technologies into India's financial infrastructure. NPCI operates several critical retail payment systems, including Unified Payments Interface (UPI), RuPay and FASTag.
Two Platforms, Different Roles
While both platforms use agentic AI, they have been designed for different purposes.
AiNxt is positioned as an enterprise-grade platform for creating, deploying and managing AI agents. It is aimed at organisations and developers looking to build AI-powered applications and workflows.
AtOM, meanwhile, is specifically focused on agent-to-agent coordination within payment ecosystems. It is designed to automate processes such as system integration, change management, partner onboarding, testing and certification.
Together, the platforms could help payment ecosystem participants reduce repetitive manual processes and accelerate the implementation of new digital payment capabilities.
What Is AiNxt?
AiNxt is an enterprise-focused agentic AI platform that allows organisations and individuals to build and deploy AI agents for different business requirements.
A key feature is its bring-your-own-model (BYOM) approach, which gives users the flexibility to work with different AI models rather than being tied to a single model provider.
The platform is designed to accommodate users with different levels of technical expertise, ranging from business users to professional developers.
AiNxt Offers Multiple Development Tools
The AiNxt suite includes AiNxt OS, AiNxt Code, AiNxt CLI and AiNxt Enterprise.
These components are designed to support different stages of AI application development, including building, testing, customisation and deployment.
The platform provides no-code, low-code and developer-oriented options, potentially allowing organisations to create AI agents without requiring every user to have advanced programming skills.
For enterprises, such flexibility could make it easier to experiment with AI applications and subsequently move successful use cases into production environments.
AtOM Targets Payment Ecosystem Automation
The second platform, Agentic Orchestration & Messaging (AtOM), has a more specialised role.
AtOM is designed around agent-to-agent orchestration, allowing different systems and participants to communicate and coordinate through automated workflows.
Within a payment ecosystem such as UPI, introducing a new feature or making a system-level change can require coordination among multiple banks, payment service providers and technology partners.
AtOM aims to streamline this process through a single workflow.
Faster Integration and Certification
One of the key applications of AtOM is the automation of integration, testing and certification.
Payment-system changes typically require extensive testing before they can be introduced across participating institutions. Manual coordination can increase the time required for implementation.
By automating parts of these processes, AtOM could help banks and payment service providers introduce and certify new UPI features more efficiently.
The objective is not simply to automate individual tasks but to connect multiple steps into a coordinated workflow.
Partner Onboarding Could Become More Efficient
Onboarding new partners into a large payment ecosystem can involve documentation, technical integration, testing, compliance checks and certification.
AtOM is designed to bring these activities into a more structured and automated process.
This could be particularly relevant as India's digital payments ecosystem continues to expand and more banks, fintech companies and payment service providers participate in interconnected payment infrastructure.
Reducing the time and manual effort required for onboarding could allow ecosystem participants to launch services more quickly.
Digital and Machine-Readable Communication
NPCI said AtOM uses machine-readable and digitally signed interactions, which can support compliance and audit requirements.
Such architecture could make it easier for participating organisations and systems to maintain consistent records of interactions and changes.
The platform is based on JSON-RPC 2.0, a lightweight remote procedure call protocol that enables structured communication between software systems.
NPCI said the architecture could potentially be extended beyond UPI to other products and payment systems.
Potential Impact on UPI Ecosystem
UPI has grown into a highly interconnected ecosystem involving banks, payment service providers, fintech companies and technology platforms.
As the number of participants and features increases, managing integrations and technical changes becomes increasingly complex.
AtOM could potentially reduce this complexity by automating communication and coordination between participating systems.
Faster testing and certification could also help new UPI capabilities reach consumers more quickly, provided the required security, compliance and operational safeguards are maintained.
AI Moves Beyond Customer-Facing Applications
The NPCI initiative also highlights an important evolution in enterprise AI.
Much of the early adoption of generative AI focused on customer-facing applications such as chatbots, search assistants and content generation. Agentic AI is increasingly being explored for end-to-end workflow automation, where AI systems can coordinate multiple tasks and interact with other software systems.
AiNxt and AtOM represent this broader approach, with AI being applied to the underlying processes that support digital financial infrastructure.
Security and Governance Remain Critical
Greater automation in payments also increases the importance of cybersecurity, authentication and governance.
Payment systems handle highly sensitive financial information and operate at large scale. Any AI-driven workflow therefore needs strong controls around identity, authorisation, audit trails, data protection and system reliability.
The use of digitally signed interactions and structured communication in AtOM is intended to support some of these requirements, but large-scale deployment will ultimately depend on robust testing and governance frameworks.
Broader Fintech Opportunity
The launch comes at a time when banks, fintech companies and payment networks are increasingly looking at AI to reduce operating costs and improve efficiency.
For financial institutions, AI-powered automation could potentially reduce repetitive technology and compliance workloads while allowing employees to focus on higher-value activities.
For India's fintech ecosystem, platforms that enable standardised AI-based coordination could also create opportunities for faster product development and integration.
India’s Digital Payments Infrastructure Enters a New Phase
NPCI's latest initiative reflects how India's digital payments infrastructure is evolving from simply enabling transactions towards increasingly automated and intelligent financial networks.
AiNxt provides a broader foundation for building enterprise AI agents, while AtOM focuses on orchestrating interactions across payment ecosystems.
If successfully implemented at scale, the platforms could help reduce manual processes, accelerate technology rollouts and improve coordination among banks, fintech firms and payment service providers.
The larger significance lies in the possibility of AI becoming part of the underlying infrastructure of India's digital payments ecosystem, rather than remaining limited to consumer-facing applications.