Press

SAN FRANCISCO, Calif. — Today an AI agent tried to buy two products from the same merchant catalog. One was approved. One was rejected.

Every layer of that transaction ran on one stack. The agent storefront. The catalog. The signed offer. The checkout. The identity. The verification. The wallet authority. Ours.

Kachyng completed an end-to-end proof of Kachyng ACom, its agentic commerce platform. Here is what happened inside that flow.

The agent discovered products through an agent-readable storefront. It compared options and selected an item. The merchant catalog issued a signed offer, with price, terms, and merchant identity bound to the transaction. Checkout verified and consumed that offer. The agent's identity was resolved, its delegation chain validated, and its wallet authority evaluated.

Then a decision: approved, rejected, or escalated for human approval.

The lower-value product was approved. The higher-value product was rejected for exceeding its delegated spending limit. Same storefront. Same agent. Same checkout path. Different authority outcome.

One thing worth being clear about: the agent storefront is not the merchant's e-commerce store. Merchants do not rebuild their site, move their catalog into a marketplace, or replatform anything. Their existing catalog becomes agent-readable. The store humans shop stays exactly as it is. Agents get their own front door. Governed, verified, controlled.

There is a lot of important work happening right now on how agents and merchants communicate. But communication standards answer one question: how do the parties talk? They do not answer the harder one: should this transaction be allowed?

Who is this agent? Who delegated it? What is it permitted to spend, with which merchants, under what terms? Who is accountable when it is wrong? That is not a messaging problem. It is a control layer, and it has to hold no matter which standards win.

That control layer needs:

Payments has seen this before. Messaging formats come and go. The durable layer is always the one that decides trust and authority. That is the layer Kachyng built. Today it ran end to end:

storefront → catalog → signed offer → checkout → identity → wallet authority → decision

“The rejected transaction is not a failure. It is proof the control layer holds.”

Huge credit to the Kachyng team, distributed across two continents, who built every layer and ran them as one. And thank you to the design partner who ran this proof with us against a real merchant catalog. More on that soon.

If you are building agent experiences, merchant catalogs, wallets, or payment infrastructure, talk to us. If you are a merchant wondering what agentic commerce means for your store, talk to us.

SAN FRANCISCO, Calif. — Kachyng Inc. today unveiled its A-Commerce platform, a purpose-built infrastructure layer designed to let AI agents operate inside the commercial world with verifiable identity, bounded authorization, compliance controls, and secure access to traditional payment rails.

A-Commerce, short for Agentic Commerce, is the emerging model in which AI agents do not stop at recommendation or assistance. They discover products, evaluate options, negotiate terms, initiate workflows, and complete transactions on behalf of people and organizations.

That shift is no longer theoretical.

Across the market, major infrastructure players are beginning to move toward agent-initiated transaction models. But while interest in agentic commerce is accelerating, the core infrastructure gap remains the same. Traditional financial systems were built around a human actor with a verified legal identity. AI agents break that assumption.

Before an AI agent can transact safely on existing rails, three questions must be answered with precision: Who is this agent? Who authorized it to act? What rules and limits govern its behavior?

Kachyng's A-Commerce platform was built to answer those questions directly.

The platform is organized around four integrated pillars.

Together, these four layers form a complete trust and execution stack between AI agents and the financial system.

“We believe the market is crossing an important line. AI agents are moving from passive assistance into real economic action. The missing layer is not more intelligence. The missing layer is infrastructure that makes those actions identifiable, authorized, governed, and executable on the rails the world already uses. That is the layer Kachyng is launching.”

Kachyng's view is that agentic commerce will not be adopted through intelligence alone. It will be adopted when businesses, financial institutions, and platforms have the confidence to let agents act without losing control, auditability, or processor flexibility.

The company built A-Commerce for that moment.

SAN FRANCISCO, Calif. — Kachyng Inc. today announced the opening of its IDX Early Access Program, inviting enterprises, financial institutions, and AI platform providers to begin integrating with the company's Identity Exchange, the cryptographic identity layer at the foundation of Kachyng's A-Commerce platform.

As AI agents take on more operational responsibility, the first problem that must be solved is identity. Before an agent can be authorized, trusted, monitored, or allowed to transact, it must first be identifiable. Today, most AI agents operate as opaque processes inside applications, workflows, or orchestration systems, with no standardized way for outside parties to verify who deployed them, who controls them, or what authority they hold.

IDX was built to change that.

With IDX, each AI agent is issued a cryptographic identity tied to its authorizing human or organization, scoped to the actions it is permitted to take, and structured so that counterparties can verify that identity across a transaction chain.

“Identity is the first gate. If an AI agent cannot present a verifiable identity, then nothing else about trust, delegation, or payment authority can be built on top of it. IDX is designed to give agents a real identity layer that works in the commercial world.”

Organizations participating in the IDX Early Access Program will receive access to core capabilities designed to help them begin integrating agent identity into their environments.

The program is intended for organizations building or deploying AI agents that will interact with financial systems, commerce systems, or transaction environments. Early participants will work directly with Kachyng's engineering team and will have the opportunity to shape the platform's path toward general availability.

Kachyng believes that standardized agent identity will become a foundational requirement for any institution or platform preparing for agent-initiated commerce.

Interested organizations can apply at www.kachyng.com/idx-early-access.

SAN FRANCISCO, Calif. — Kachyng Inc. today introduced KYA, or Know Your Agent, a compliance framework designed to help financial institutions, payment providers, and commerce platforms verify, authorize, and monitor AI agents before those agents are allowed to initiate activity on traditional payment rails.

The rise of AI agents creates a direct compliance problem for the financial system. Existing frameworks such as KYC, or Know Your Customer, and AML, or Anti-Money Laundering, were built for human beings and legal entities. They assume the party initiating a transaction can be tied to a person or company through documents, beneficial ownership records, and established risk controls.

AI agents do not fit neatly into those categories.

An AI agent may be acting on behalf of an individual, a business, or an internal workflow, but the agent itself is not a human customer. That creates a new trust gap. If an AI agent initiates a payment, subscription renewal, procurement request, or account-level action, financial institutions need a structured way to determine whether that agent is legitimate, who stands behind it, and whether it is operating within authorized limits.

KYA was designed to solve that problem. Kachyng's KYA framework works in three stages.

“KYA is our answer to a simple but critical question. If an AI agent is going to act inside the financial system, what is the compliance model for that actor? We believe the industry needs a framework that extends existing controls to agents without pretending agents are just another type of human customer.”

Kachyng designed KYA to work with existing compliance systems rather than replace them. The goal is to help institutions extend proven controls into a world where autonomous software begins participating directly in financial activity.

As agentic commerce moves from theory into deployment, Kachyng believes compliance infrastructure for non-human actors will become a core requirement for banks, processors, merchants, and enterprise platforms.

SAN FRANCISCO, Calif. — Kachyng Inc. today introduced A-Commerce, short for Agentic Commerce, a new infrastructure category built for a world in which AI agents do not just recommend or assist, but actually carry out commercial work on behalf of people and organizations.

AI is rapidly moving beyond chat, search, and summarization. Agents are beginning to compare offers, negotiate terms, renew services, manage subscriptions, place routine orders, and handle purchasing tasks. But while the intelligence layer has advanced quickly, the transaction layer has not. Existing financial infrastructure was built around one core assumption: that every transaction begins with a human being who can be identified, authorized, and held accountable.

That assumption does not hold when the actor is an AI agent.

Before any AI agent can be trusted to initiate a transaction on traditional rails, three questions must be answered clearly and deterministically: Who is this agent? Who gave it authority to act? What limits govern what it is allowed to do?

Kachyng calls the infrastructure required to answer those questions A-Commerce. Kachyng's platform is built around four integrated pillars.

Together, these four layers form a complete trust architecture between AI agents and the financial system.

“We believe the next major shift in commerce is not simply better software for humans. It is the emergence of AI agents as real commercial actors. But that future only works if agents can be identified, governed, constrained, and connected to execution. That is the layer Kachyng is building.”

Kachyng believes A-Commerce will become foundational infrastructure as businesses begin deploying AI agents into real operational and financial workflows. The company's mission is to make agent-initiated commerce possible without sacrificing compliance, auditability, control, or compatibility with the financial systems already in place.