# Catena — Full content
> Catena works at the intersection of AI agents, financial systems, identity, and stablecoins. Our goal is to realize the potential of AI to increase prosperity for everyone.
Full markdown content of every public Catena blog post, sorted newest first. Generated at build from content/posts/.
---
# x402 on Catena: Dollar Accounts, Controls, and Evidence
Published: 2026-08-04
URL: https://catena.com/blog/x402-on-catena-dollar-accounts-controls-and-evidence
Summary: How Catena brings x402 payments into governed banking: customer-defined controls, dollar-account funding, and linked evidence for every agent transaction.
TL;DR: Catena supports x402 natively with customer-defined controls: agents pay for services straight from dollar accounts, within policy and spending limits, and every payment lands as a first-class ledger transaction with linked evidence.
---
AI agents increasingly need to buy things to do their jobs: API access, datasets, compute, browser sessions, and specialized research. However, these payments look different from traditional business purchases. They tend to be task-oriented, pay-per-use, and made at machine volume. Accordingly, a handful of new payment protocols have been developed to serve this burgeoning agent activity.
One of these protocols is x402.
When we added native x402 support to our banking platform, we focused on applying customer-defined controls and existing business processes around its use. As a result, we have unlocked something of real value for the early customers on our platform: the opportunity to leverage paid, agent-driven tasks safely at scale.
## Customer-controlled authority
With standalone, bare-bones agent payments, signing authority for the transactions typically sits quite close to the agent runtime. That can increase the potential impact of prompt injection, compromised dependencies, or other hostile inputs reaching the process.
On Catena, an agent always first authenticates before submitting a payment intent. The intent is evaluated against customer-defined policies, spending limits, approved providers, and approval requirements. The signing architecture is designed so that neither the agent nor Catena can move customer funds alone. Catena enforces the customer's instructions; it does not replace the customer as the source of financial authority.
We previously wrote about moving policy enforcement into the signing layer itself, and x402 has been an excellent demonstration of how this multi-layered approach enables agent operators to use payment protocols, such as x402, with bounded authority, layered controls, and a clear record of what happened.
## Flexible dollar accounts
Many agent payment protocols, including x402, leverage stablecoins as a faster, programmatic way of exchanging value. However, not all businesses have ready access to these stablecoins. Setting up accounts with the relevant providers can be difficult, leaving companies running payments out of two separate places.
Catena lets businesses pay for x402 services straight from their dollar accounts. If a service settles in a supported stablecoin, Catena handles the conversion with the customer's authorization, and the payment runs through the same policy, approval, signing, and evidence controls as any other transaction. That means a business does not need to become operationally stablecoin-native before its agents can access services offered through agentic payment protocols.
## Evidence beyond settlement
A settlement response can show that a payment occurred, but businesses also need the surrounding operational context: which agent initiated it, what the customer authorized, which policy was evaluated, whether human approval was required, and how the result appears in the business's records.
Catena preserves that context around all paid-tool commerce. The agent proposes an action; policy evaluates it; the action is approved, held for human review, or blocked; an approved action executes and settles; and the policy, approval, settlement, and evidence context remain linked afterward.
Every payment is a first-class ledger transaction, including the $0.001 ones, with the original 402 challenge preserved alongside it. Operators can review linked evidence, inspect activity across supported payment methods, export reports, and investigate or disposition operational exceptions without rewriting the underlying transaction history.
## Autonomy within clear limits
A diligence agent might purchase a company-registry lookup, a KYB report, and selected records from an approved data provider. A coding agent might pay for a test environment, browser session, evaluation run, or GPU job.
At machine volume, the answer is not to approve every small payment manually. The business needs to be able to pre-authorize routine purchases within explicit provider, amount, and velocity limits, while unusual amounts or unapproved providers are held for human review.
Agent payments can offer enormous advantages over traditional payment rails, but only when used within a clear and reliable policy system. Our approach to x402 reflects our approach to agent payment protocols more generally, including Machine Payments Protocol or Mastercard's Agent Pay for Machines: as one of many options to be leveraged within a durable and consistent financial product.
Catena is currently in private access. [Join the waitlist](https://app.catena.com) to explore governed agent commerce (including x402 workflows) with test funds in a secure sandbox and move to production when your policies are fully tested.
# Binding Policy to Money
Published: 2026-06-23
URL: https://catena.com/blog/binding-policy-to-money
Summary: Catena now enforces policy at the signing layer inside a TEE, so funds can't move on the agent's say-so, on Catena's alone, or without your approvals.
TL;DR: Catena now enforces agent policy at the signing layer itself, inside a Trusted Execution Environment (TEE) built with Turnkey. Funds can't move on the agent's say-so alone, on Catena's alone, or without the approvals you require.
---
Today we're adding a new layer to the Catena control plane: policy enforced at the signing layer itself, inside a Trusted Execution Environment (TEE). It's the strongest control we've shipped for agentic money movement thus far, and the property that makes it matter is simple. At this layer, funds can't move on the agent's say-so alone, on Catena's alone, or without whatever approvals you require. The agent can't go rogue, and neither can we.
[Turnkey](https://www.turnkey.com/) is our technology partner for this new TEE-based policy enforcement, strengthening the infrastructure that enables businesses to operate agents at scale.
## From evaluation to enforcement
Catena's core policy engine turns an operator's rules into capabilities and limits. Which accounts an agent can touch, which counterparties, how much, how often, and what needs a human to sign off—all of this forms a unique policy, governing each and every agent's intention before anything executes.
Until recently, an agent's policy was evaluated at two levels: at the agent level, where its identity is defined and scoped, and at the application level, where Catena checks an agent's permissions across different rails and features. Now, it's also enforced at the moment a transaction is signed. This approach uses a secure enclave (TEE) that the application cannot reach into or talk its way around, so that policy moves with money instead of sitting outside money movement as a suggestion.
Think of it like a vault that opens only when the required keys turn together: the agent's request, the operator's policy, and any approvals you've set. There's no master key to misuse, not even Catena's, which means breaking the rule isn't something you catch after the fact—it isn't possible in the first place.
The process of provisioning agent identity stays simple, leveraging well-established authentication methods to delegate sending authority, but it now gets reinforced by real cryptographic signatures, securing the agent's permissions without the need for key management.
## Layered by design
The job of Catena's policy engine is to answer a few essential questions about every agent: who is it, what is it allowed to do, and can you prove what it did afterward.
By enforcing policy directly in the signing layer, we now strengthen the most important consequence of the "who" question for agents: authorization of money movement. This is the third layer in a multi-stack control plane: customers set guardrails for what agents can attempt; application policy surfaces the appropriate actions for those attempts; the signing layer checks permissions to move funds; and the fourth layer, now possible with the technology introduced by the TEE, encodes customer guardrails directly into smart contracts.
Together, these layers produce a system designed for long-term trust and reliability. This is how you let agents move money safely at scale.
Get started and apply for private access today at [app.catena.com](https://app.catena.com).
# Catena Supports Mastercard Agent Pay for Machines
Published: 2026-06-10
URL: https://catena.com/blog/catena-supports-mastercard-agent-pay-for-machines
Summary: Mastercard has launched Agent Pay for Machines, and Catena is among the initial participants bringing identity, policy, and governance to agentic payments.
TL;DR: Catena is one of the initial participants in Mastercard Agent Pay for Machines, a new service for permissioned, machine-speed agent payments. Catena adds the control plane: identity, policy, approvals, spending limits, and auditability wherever agents move money.
---
Today Mastercard [launched](https://www.mastercard.com/us/en/news-and-trends/press/2026/june/mastercard-launches-agent-pay-for-machines.html) [Agent Pay for Machines](https://www.mastercard.com/us/en/business/artificial-intelligence/mastercard-agent-pay/agent-pay-for-machines.html), a new service for payments that are permissioned, orchestrated, and settled at machine speed across its global network. AI agents acting on behalf of people and businesses are beginning to buy compute, data, API access, and services continuously, often turning one human instruction into a chain of automated transactions across providers. Agent Pay for Machines gives those transactions credentials, programmatically enforced spending limits, and guaranteed settlement across payment types, from cards to stablecoins. We’re glad to be among the initial participants supporting the launch.
Our view is that payment rails alone are not enough. AI agents moving money need real controls: identity, policy, approvals, spending limits, counterparty rules, and auditability. That is why we are building a banking and governance platform that enforces those controls across every meaningful rail, protocol, and on-chain contract. For the people and businesses who deploy agents, Catena is a single control plane that applies one consistent set of policies wherever money moves.
The future of agentic commerce will not run on one network or one rail, and that diversity is healthy. With Catena, agents can move money safely over multiple rails, currencies, and networks as needed, with clear constraints and audit trails. As part of the Agent Pay for Machines ecosystem, we’ll help validate priority use cases, support common rules, and keep doing what we do wherever agents transact: make agentic payments safer, more accountable, and more practical for customers.
If your agents are starting to move money, apply for Catena access at [app.catena.com](https://app.catena.com).
# Our Banking and Governance Platform for AI Agents is Open
Published: 2026-05-14
URL: https://catena.com/blog/banking-governance-platform-for-ai-agents-open
Summary: Private access for the Catena banking and governance platform is open. We've also raised a Series A and applied for a national trust bank charter.
---
# Our Banking and Governance Platform for AI Agents is Open
Agents are trying to move money. In OpenClaw, in Claude Cowork, in ChatGPT and Codex, in dozens of agent platforms, they are attempting to hold funds, negotiate, buy and sell, and earn yield on idle assets on behalf of the people and businesses who deploy them. Mostly, they can't. The financial system was built for humans with wallets and paper forms, not for software acting under policy on someone else's behalf.
That is the gap Catena is closing. Today, we are announcing three things.
## Private Access is open
Catena's Private Access program is now open. Select businesses are already using our services across two core areas: trust and compliance, including agent identity, reputation, and governance; and financial operations, including fiat and stablecoin liquidity, risk management, and yield generation. More services are rolling out in the coming months.
Catena is one platform with two surfaces, by design. For agents, it is a banking platform. They can get verified identities, open accounts, hold dollars, send and receive payments, and transact on modern rails, including stablecoin and card rails. Agents do not need banking websites. They need atomic, programmable services they can assemble for specific tasks.
For the humans and businesses who deploy them, Catena is a control plane. Human operators set policy, approve activity, cap spending, restrict counterparties, audit every movement, and halt agent activity when necessary. This governance layer applies to both Catena-provided accounts and non-custodial accounts controlled by agents and their operators. We will expand these capabilities to include governance over third-party provider accounts that our customers want to access using agents in the coming months.
The principle is simple: not autonomy without accountability, but identity with accountability. Agency without surrender.
## Our OCC charter application is in progress
We are building a regulated financial institution purpose-built for agents and their operators. As we enter a new era of agentic commerce, we are committed to doing things right. We have filed for a National Trust Bank charter and will continue to work with federal regulators through the chartering process.
## $48 million in total financing
We are also disclosing a $30 million Series A, co-led by a16z crypto and Acrew, with participation from Breyer Capital, General Catalyst, QED, Oak HC/FT, Fin Capital, Coinbase Ventures, and IDG Capital. That brings our total financing to $48 million.
We are using this capital to build the technology platform, obtain the licenses, and hire the people required to create a trusted platform for AI agents and their operators at scale.
## What happens next
Agent identity, policy enforcement, payments, yield, liquidity, and clearing are building blocks. The services agents need will grow as agents themselves get more capable.
None of this works without the assurance of safety. That is what we are building. If you are running agents that touch money, or are about to, come build with us.
Apply for [Private Access](https://app.catena.com). Join our [Discord community](https://discord.com/invite/catenacommunity), reach out on [X](https://twitter.com/catena_labs), and contact us at [hello@catena.com](mailto:hello@catena.com).
---
# What Agents Need Before They Handle Real Money
Published: 2026-02-06
URL: https://catena.com/blog/what-agents-need-before-they-handle-real-money
Summary: Today our AI agent banking system began managing a portion of our own corporate funds—live agents managing real money.
---
There are 1.5 million agents transacting on Moltbook right now. Depending on who you ask, this is either the early singularity, a dumpster fire, or 17,000 humans puppeting bots. Wiz Research found 341 malicious skills on ClawHub stealing credentials. Karpathy called it what it is.
Here's the thing: it doesn't matter which take is right. The infrastructure problems are the same regardless. Agents that move money need guardrails that actually work — not guardrails that politely suggest limits and hope the application layer behaves.
We've been building agentic treasury and banking infrastructure at Catena — with the same focus on compliance and institutional trust that I learned building USDC at Circle — for a while now. Quietly, mostly. But the OpenClaw moment made it clear that these problems are suddenly relevant to a much larger audience than we expected, much sooner than we expected. So we recorded a quick raw demo.
## The demo
The video is six unpolished minutes, no production, just direct content. I'd recommend watching it before reading the rest of this, but here's the short version.
We show an AI agent with its own USD account, powered by Turnkey's wallet infrastructure, holding stablecoin on Base. Users see human-readable labels — "Visa •••4532", "Chase Checking •••7891", "Catena Treasury Agent" — not raw blockchain addresses. Then we run it through a series of transactions that exercise different policy paths:
**Auto-approved rebalance.** The agent sends $10 to a treasury agent. Policy says under $50/day to known treasury addresses is fine. It executes instantly. No human in the loop.
**Approval required.** The agent tries to pay $8 on a Visa bill. Policy requires human approval for external payments. An approval card appears. The user approves, and it goes through.
**Rejection.** The agent tries to withdraw $25 to a Chase checking account. Same approval flow. The user rejects. Funds don't move. And — this is the important part — even if the AI application layer were completely compromised, Turnkey's policy engine still wouldn't allow a signature without that approval.
**Hiring another agent.** The user asks to hire a Meridian research agent. The system checks the agent's decentralized identity, pulls its reputation score (87/100, 142 attestations), and auto-approves because the score meets the configured threshold. Agent-to-agent commerce with identity verification built in.
**Standing authorization.** The user delegates ongoing spending authority — $20/month on Meridian reports. Two-layer enforcement applies to recurring authorizations too. Then we check how the treasury agent is actually allocating capital across a portfolio view with APY breakdowns.
The whole thing runs in two surfaces: Claude Desktop using MCP with rich interactive cards, and OpenClaw over WhatsApp as a text-based skill. Same agent, same policy enforcement, different interfaces. The surface changes. The security model doesn't.
## The core idea: two layers, not one
Most agent frameworks treat policy as an application-layer concern. The AI decides whether a transaction should happen, checks some rules in code, hopefully works within prescribed guardrails, and proceeds. This is fine for demos. It is not fine for real money.
The problem is straightforward: application-layer policy is only as secure as the application. If someone compromises the server, jailbreaks the model, or finds a bug in your policy-checking code or guardrails framework, the money moves. You've built a lock out of suggestions.
What agents actually need are two layers:
**Layer 1 is intelligence.** This is the application layer — the part that answers the questions you'd want answered before any money moves. Who controls this agent? Are they a verified entity? What's their track record? You can see this in the demo: before the treasury agent pays another agent for research services, it resolves their identity, checks their reputation score, and evaluates whether they meet the policy threshold. An agent with a verified owner, a score of 87/100, and 142 attestations clears. An unverified agent with a dispute flag doesn't. This is the kind of automated standards-based trust infrastructure that the agentic economy needs — not platform-specific API keys, but portable, verifiable identity that works across any agent framework.
**Layer 2 is enforcement.** In this example, this part runs in Turnkey's secure enclave. Has the required approval been obtained? The enclave signs the transaction only if every policy condition is met. This isn't running in our application code. It's running in hardware that neither we nor Turnkey can tamper with after deployment.
Intelligence without enforcement is just prompt suggestions. Enforcement without intelligence is just a dumb access list. You need both.
Even if our entire backend is compromised, the enclave won't sign transactions that violate policy. That's not a promise — it's math.
## How the custody actually works
This is worth being specific about, because "non-custodial" gets thrown around loosely.
The signers are the customer (via passkey on their device), the customer's agent, and the Catena treasury agent. Turnkey enforces a quorum — 2-of-2 or 2-of-3 depending on the configuration — so no single party can move funds alone.
Different combinations handle different scenarios. For routine operations within policy — say, a small rebalance into an approved vault, with spend limits and recipients guided by human-defined policy — the agents can co-sign automatically. No human approval needed. The default policy requires the customer's passkey for sensitive operations like withdrawals — but the customer defines those policies, not us. And in a 2-of-3 setup, the customer and their agent can reach quorum without Catena entirely. There's always an exit path.
The policies governing all of this live both in [ACK-ID](https://www.agentcommercekit.com/ack-id/introduction) verifications (including integration with ERC-8004) as well as inside Turnkey's secure enclave. The policies define which identities are approved, what amount thresholds or reputational thresholds trigger additional approval, and which operations the agents can handle on their own. Beneath the agent application layer, the enclave evaluates every transaction against its part of these rules before signing. Agent code can't override it.
The architecture reflects problems we've been working on together: how do you give agents real financial capability without creating a new category of catastrophic failure mode?
## Why now
Six months ago, agent-to-agent payments were a theoretical concern. Now there are seven-figure sums moving through agent networks with, in many cases, no policy enforcement at all. The gap between what agents can do and what the infrastructure supports them doing safely is widening fast.
We're showing what we're building because the people building agents right now — in the OpenClaw ecosystem and beyond — are going to run into these problems immediately if they haven't already. We love the vision and are optimistic, but the 341 malicious ClawHub skills are not an anomaly. They're the beginning. We have to get this right, because getting it wrong has dire consequences.
This is a glimpse of where we're headed, not a finished product. We're building financial infrastructure that agents and their owners can trust — and we'd rather get it right with the community than in isolation. If any of this resonates, or if you think we're wrong about something, let us know.
# Being Client Zero
Published: 2025-12-18
URL: https://catena.com/blog/being-client-zero
Summary: Today our AI agent banking system began managing a portion of our own corporate funds—live agents managing real money.
---
Today our AI agent banking system began managing a portion of our own corporate funds—live agents managing real money.
This isn't a demo environment or testnet deployment. It's our actual capital, managed by the financial infrastructure we're building for the agentic economy.
## Trust First
Before we ask anyone else to trust their treasury to our AI agents, we need to prove the system works with our own money. That's what being Client Zero means—subjecting ourselves to the same standards, the same guardrails, and the same accountability we extend to customers.
Every agent action operates within the strict policy boundaries we've defined. Every transaction is cryptographically verified. Human oversight remains embedded in policy and key decision points.
## Why This Matters
Traditional banking infrastructure wasn't designed for AI agents as economic actors. Building a financial system that safely enables agents to transact requires rethinking identity, authorization, compliance, and risk management from the ground up.
We're building that infrastructure. This is just the beginning.
Follow us on [X](https://twitter.com/catena_labs) for updates as we continue to build.
# How AI Uses Money: The Case for Open Standards in Agent Commerce
Published: 2025-11-07
URL: https://catena.com/blog/agent-commerce-standards
Summary: Standards in Agent Identity & Commerce
---
_This post summarizes our Web Summit 2025 presentation: "How AI Uses Money" with links for further research._
---
In 1995, e-commerce was ready to explode, but it was stalled. We had browsers and websites, but no trust. The creation of SSL/TLS—an open, cryptographic trust layer—wasn't just an update; it unlocked a trillion-dollar economy.
Today, we are at the exact same moment with AI. AI agents are ready to transact... but they're stalled. We are missing the trust standards for AI commerce.
## The Problem: Fragmentation Before We Begin
Today, merchants face an impossible choice: allow "bots" to transact and risk fraud liability, or block them entirely and lose revenue. Without interoperable standards for agent identity and authorization, the default is to block.
Meanwhile, legacy payment infrastructure wasn't designed for machines. Agents need instant settlement, global connectivity, and programmable guardrails, but they're stuck with systems built for human speed, business hours, and manual error handling.
Instead of collaborating on shared solutions, companies and teams are currently building competing solutions, or forking and fragmenting nascent proposed standard solutions. Multiple new protocols launched in 2025 (AP2, ACP, x402, ATXP, KYAPay, etc.). AI providers, fintech platforms, and new payment startups are creating proprietary systems that do not interoperate.
## What's Missing
For humans, we agreed upon trust and payment standards long ago:
- **Identity:** KYC/KYB regulatory standards and practices (with government and network rules)
- **Authorization:** OAuth 2.0 and related extensions for delegated access
- **Policy:** Card networks and banks enforce limits, authentication (e.g., 3-DS), and fraud controls
- **Payments:** Card/bank rails (PCI/EMV/ACH/RTP), secured over TLS/HTTPS during transport
- **Audit:** Comprehensive reporting and accountability (network + PSP + merchant logs)
For agents? Nothing. Every field is blank. Different companies are reinventing different wheels.
## History Shows Open Standards Work
Open standards unlock trillion-dollar economies:
- **TLS** secured the web
- **TCP/IP** connected the internet
- **SMTP** made email universal
- **SMS** made every phone interoperable
- **Shipping containers** revolutionized global trade
These succeeded because everyone — including market competitors — benefited from interoperability. Open compounds value. Closed caps it.
## What the Ecosystem Needs
Four foundational components must work together:
### 1. Agent Identity (KYA: Know Your Agent)
Universal cryptographic proof that works across all protocols. Agents need verifiable credentials that establish identity, delegation, and authorization.
### 2. Authorization, Policy & Safety
Verifiable delegation chains, programmable guardrails, and safety rules that travel with the transaction. When an agent acts, we need to know: who operates it, under what constraints, and what it's allowed to do.
### 3. Transparency & Accountability
Immutable audit trails. When something goes wrong, liability must be clear. Who authorized what, when, and why? This isn't optional, it's table stakes for any system handling real money.
### 4. Payment Protocol Interoperability
We need standardized protocols that let agents pay for resources, pay humans, get paid, and transact with each other, regardless of the underlying payment infrastructure.
All of this must be built on **open standards**. No single owner, no walled gardens.
---
## What Catena Labs Is Building
Catena Labs is partnering with AI providers, agent developers, fintechs, stablecoin issuers, and banks to define open trust standards for the agent economy. We’re building an AI-native financial institution on interoperable protocols so agent commerce is open, auditable, and safe at internet scale.
---
## Join the Conversation
The window for building open standards is narrow. Early architectural choices compound—we've seen this with the internet, email, and mobile networks. The decisions we make in the next 12-18 months will determine whether we build interoperable protocols or fragment into incompatible silos.
**Want to collaborate?**
**Email:** [hello@catena.com](mailto:hello@catena.com)
**Twitter:** [@catenalabs](https://x.com/catena_labs)
**LinkedIn:** [Catena Labs](https://linkedin.com/company/catenalabs)
---
## References: Who’s Building What
### Core Agentic Protocols (Comms, Tools, Commerce)
- **[Agent-to-Agent (A2A) Protocol](https://a2a-protocol.org/latest/)** — Google's proposal for agent discovery and communication.
- **[Model Context Protocol (MCP)](https://modelcontextprotocol.io)** — Anthropic-originated de facto standard for connecting agents to tools, data, and services.
- **[Cloudflare Agents SDK](https://developers.cloudflare.com/agents/)** — SDK + edge runtime for deploying durable agents in Cloudflare.
- **[CrewAI](https://www.crewai.com)** — Multi-agent orchestration framework (Python).
- **[LangChain](https://www.langchain.com/)** — Agent framework (LangGraph, LangSmith) for building production LLM apps.
- **[LlamaIndex](https://www.llamaindex.ai)** — Agent/RAG framework with document/workflow tooling.
- **[Vercel AI SDK](https://ai-sdk.dev/docs/introduction)** — TypeScript toolkit for apps/agents on Node/Next.js.
### Agent-Facing Payment & Commerce Protocols
- **[Agent Commerce Kit (ACK)](https://agentcommercekit.com)** - Patterns for standards-based agentic identity and payments curated by Catena.
- **[Agentic Commerce Protocol (ACP)](https://developers.openai.com/commerce/)** — OpenAI + Stripe’s proposal for agent-to-merchant checkout.
- **[ACP Spec Hub](https://agenticcommerce.dev)** / **[GitHub](https://github.com/agentic-commerce-protocol/agentic-commerce-protocol)** — ACP specification and reference implementations.
- **[AP2 — Agent Payments Protocol](https://ap2-protocol.org/)** — Google's proposed protocol for secure agent payments across methods.
- **[x402 Protocol](https://www.x402.org/)** — Machine-native payments using HTTP-402 and stablecoins.
- **[ATXP (Agent Transaction Protocol)](https://circuitandchisel.com/)** — Circuit & Chisel’s protocol for multi-party agent payments.
- **[Skyfire's KYAPay](https://skyfire.xyz/)** — Identity-linked agent payments from Skyfire.
- **[Payman](https://paymanai.com)** — AI-native payments enabling agents to pay humans with policy controls.
### Network / Provider Initiatives
- **[Visa Intelligent Commerce](https://developer.visa.com/capabilities/visa-intelligent-commerce)** — Visa's agentic commerce toolkits and developer APIs.
- **[Mastercard Agent Pay](https://www.mastercard.com/us/en/business/artificial-intelligence/mastercard-agent-pay.html)** — Mastercard's trust and payments stack for agent experiences.
- **[Stripe Agent Toolkit](https://docs.stripe.com/agents)** — SDKs and patterns to add Stripe capabilities to agents.
### Durable Orchestration & Workflows
- **[Inngest](https://www.inngest.com/ai)** — Durable, event-driven workflows and queues for agents.
- **[Trigger.dev](https://trigger.dev/)** — Long-running jobs, retries, and observability for AI agents.
- **[n8n](https://n8n.io/ai-agents/)** — Visual workflows and AI nodes with 400+ integrations.
- **[Zapier AI Actions](https://actions.zapier.com/)** — Expose 6k+ app actions to agents.
### Foundational Identity & Authorization Standards
- **[Decentralized Identifiers (DID) — W3C](https://www.w3.org/TR/did-core/)** — Identifier + document standard for decentralized identity.
- **[Verifiable Credentials Data Model v2.0 — W3C](https://www.w3.org/TR/vc-data-model-2.0/)** — Cryptographic credential format and ecosystem.
- **[OAuth 2.0 — IETF RFC 6749](https://datatracker.ietf.org/doc/html/rfc6749)** — Standard for delegated authorization.
# Catena Partners with Circle to Support Arc: Built for Agentic Finance and Commerce
Published: 2025-10-28
URL: https://catena.com/blog/catena-partners-with-circle-to-support-arc-built-for-agentic-finance-and-commerce
Summary: We are excited to announce Circle has partnered with us to bring AI banking and financial services to the new Arc L1 blockchain, which has been designed from the ground up for stablecoins and payments.
---
Today, we're excited to announce that Circle has partnered with us to build support for agentic commerce and finance on Arc, their new Layer-1 blockchain designed specifically for payments and stablecoins. As an AI bank focused on providing financial infrastructure for AI agents, our approach is to work across a range of blockchains and legacy payment rails to address our customers' needs. However, we see the clear demand for new L1 chains that are purpose-built for payments. Arc addresses this need with the kind of foundational technology the agent economy needs.
## **Why Arc Matters for Agents**
AI agents need to transact at machine speed with predictable costs and instant settlement. Arc delivers this through:
- **Stablecoins as Native Gas**: Agents pay transaction fees directly in USDC, eliminating the complexity and volatility of managing tokens to pay gas fees. No more converting between assets just to execute a payment.
- **Deterministic Sub-Second Finality**: Transactions settle instantly with certainty. Agents don't wait days for settlement or deal with probabilistic finality—they know immediately when a payment is complete.
- **Purpose-Built for Payments**: Unlike general-purpose chains adapted for payments, Arc was designed from the ground up for the movement of money. This means lower costs, higher reliability, and infrastructure optimized for the transaction patterns agents will generate.
## **Agent-Native Money on Agent-Native Rails**
We've long believed stablecoins are AI-native money—programmable, global, and operating at software speed. But even the best money needs the right rails. Legacy payment systems impose human-speed constraints, geographic restrictions, and friction incompatible with semi-autonomous agent workflows.
Arc changes this. When an agent needs to pay for API access, purchase training data, or settle a B2B invoice, it can transact directly using stablecoins without navigating traditional banking hours, currency conversions, or multi-day settlement windows.
As a new AI Bank centered on stablecoins, we're enabling a future where agents are semi-autonomous participants in the economy. Agents need a reliable, efficient, compliant, secure, and predictable financial infrastructure. Arc meets this standard.
Arc represents another critical building block for the new agentic finance stack. These new infrastructure and services, built from the ground up for AI are unlocking new opportunities for founders and developers to build innovative new products.
## **Building Together**
We're excited to collaborate with Circle to build on Arc and create a new generation of financial services built for AI with AI.
Circle is the leader in regulated stablecoins. They are driving the adoption of USDC as a global dollar-based payment instrument. USDC has tremendous momentum in the market today. The USDC ecosystem will leap forward with the adoption of Arc.
At the same time, we’re embracing the whole ecosystem of blockchain and legacy technologies that will need to be integrated into the agent economy and the agentic finance stake. We’re excited to see more teams drive blockchain innovations, expand the potential of stablecoins, and accelerate adoption.
# The Agentic Finance Stack Protocols Part 1: Payment Rails
Published: 2025-10-17
URL: https://catena.com/blog/agentic-finance-stack-protocols-part-1-payment-rails
Summary: Engineers and product leaders building agentic systems that will eventually use money are well served by understanding the basics of traditional payment rails and how these will be changed by the emerging AI economy.
---
Every AI agent that buys, sells, or pays needs to move money. But modern payment systems were built for humans, not autonomous code. Let's explore how money actually moves today—and why it matters for agentic commerce.
In this post, we’ll explore the first layer of the [Agentic Finance Stack](https://catena.com/blog/the-a-commerce-stack-building-the-financial-capabilities-for-ai-agents): the foundational payment protocols and rails that move money in today’s financial system.
There are many layers of protocols, networks, and systems that enable modern payments. We won’t cover all of them here. Instead, we’ll focus on the key technologies that teams building agentic systems need to understand.
Any payment between AI agents, or between agents and other systems (like MCP servers), must move across a payment rail. While we expect stablecoins to become the primary payment instrument for agentic systems, they won’t be the only ones. Understanding how traditional rails work—and how they differ from stablecoin systems—will help AI builders design reliable, compliant, and interoperable agentic payments.
---
## **Key Payment Concepts**
Most traditional (non-stablecoin) payment systems share **five foundational steps** in the flow.
_Figure generated by ChatGPT 5 Auto._
These steps may not always occur in strict order. For example, clearing and posting may happen before settlement. Among these, settlement is often the most critical step—this is when funds are finally transferred between financial institutions to discharge obligations.
Another essential concept is liquidity—the availability of funds in the right form and location at the moment they’re needed to settle a transaction. If liquidity is unavailable, even valid payment instructions can’t be completed. Liquidity management is central to payment operations and will be revisited in later posts.
## **Bank-to-Bank Payments**
Payments between banks form the backbone of the global financial system. Many of the systems still in use today were established decades ago, but remain critical underpinnings of modern money movement.
While our long-term goal is to replace legacy rails with stablecoin-based systems and new protocols designed for AI agents, these older systems will remain relevant for years to come.
### **Wires**
Wire transfers date back to the 19th century, when banks used telegraph messages and physical couriers to move funds. Today, modernized systems like Fedwire (U.S.) and SWIFT (international) enable secure, high-value transfers.
- **Fedwire** settles transactions between banks’ accounts at the **Federal Reserve**, enabling **fast and final settlement** compared to most other traditional rails.
- Wires are **relatively fast**, support **large-value** transactions, and offer **finality** (they can’t be reversed by the sender).
- However, they’re **bank-to-bank only**, operate on **limited hours**, carry **high per-transaction costs**, and include **minimal message data**.
While agents could technically initiate wire transfers, they are ill-suited for real-time, high-frequency, or low-value use cases that agentic systems typically need.
### **ACH (Automated Clearing House)**
The ACH network, launched in the 1970s as an electronic alternative to paper checks, is the most widely used U.S. payment system. In 2024, ACH processed over 31 billion transactions, moving more than $80 trillion.
- ACH transactions are **batched** and **settled several times per day** through the Federal Reserve.
- Typical settlement occurs within **1–2 business days**, though **Same Day ACH** now enables faster transfers in many cases.
- ACH supports both **credit (push)** and **debit (pull)** payments, with **low costs** and **broad adoption**.
However, ACH is not instant, and transactions can be reversed for a period. These characteristics make it unsuitable for millisecond responsiveness or always-on agentic use cases. It may serve as an underlying layer in hybrid systems, but is unlikely to be the primary interface for AI agents.
### **Real-Time Payments (RTP)**
To modernize bank transfers, new real-time payment systems have emerged—such as The Clearing House RTP and FedNow.
- These systems operate **24/7**, offer **instant or near-instant settlement**, and are designed for **finality** (though limited recall mechanisms exist).
- They support **low-cost**, **irreversible**, and **account-to-account** payments.
- However, they remain **bank-to-bank systems**, often **capped** in transaction size (e.g., $500K), and require integration with bank partners.
RTP systems are a step toward continuous, programmable money movement, but their current design and access constraints make them challenging for autonomous AI agents to use directly.
## **Consumer and Business Transactions**
Above the bank rails sit **payment networks** and **services** that enable everyday consumer and business transactions. These systems **depend on bank rails** under the hood but add their own protocols and features.
### **Cards**
Credit and debit cards are **ubiquitous** in global payments, connecting billions of consumers and merchants.
While their **global reach** is unmatched, card networks introduce challenges for agentic systems:
- **Reversibility** (chargebacks) undermines transaction finality
- **High fees** and **multi-day settlement** increase costs
- **Cardholder data** requirements create **security and compliance** hurdles for non-human agents
As a result, card networks are **poorly suited** for autonomous AI transactions.
### **Peer-to-Peer (P2P) Systems**
Platforms like **Venmo** and **Cash App** offer **instant, low-cost, irreversible** transfers within **closed networks**.
While convenient for human users, their **closed-loop** nature (you can only pay other users of the same platform) makes them **inflexible** for agents that need to transact across ecosystems.
### **Remittance Networks**
Cross-border remittance providers such as **Western Union** and **MoneyGram** allow near-instant transfers between countries by maintaining **liquidity pools** in each region.
For example, when sending USD from the U.S. to MXN in Mexico, the provider pays out **immediately** from local reserves, then periodically **rebalances** across borders.
This **pre-funded model** enables speed but at the cost of **high fees** and **closed access**. These systems could eventually expose **APIs** for agentic use, but **integration barriers** and **cost structures** remain significant.
---
## **Stablecoins**
Stablecoins like USDC represent a new class of payment instrument with properties well-aligned to agentic commerce:
- **On-chain settlement**: Transactions are recorded and finalized on **public blockchains**, independent of central banks.
- **Instant or near-instant settlement**, limited only by blockchain confirmation times
- **Low fees** relative to traditional rails
- **Programmability** via **smart contracts**, enabling autonomous, conditional, or recurring payments
- **Irreversibility**, ensuring finality once confirmed
- **Fully digital**, with machine-native interfaces suitable for AI execution
However, stablecoins also introduce new considerations:
- **Technical integration** requires blockchain infrastructure and key management
- **On/off-ramp friction** remains, though improving
- **Irreversibility** can be a drawback in cases where refundability is desirable
- **Compliance** obligations still apply—companies transacting with stablecoins must meet **AML/CFT** requirements, often requiring licensed partners or custodians
While stablecoins are not yet a complete replacement for traditional rails, they represent the most promising foundation for autonomous, programmable, and global agentic payments.
## **Additional Considerations for Agentic Systems**
To design effective payment systems for AI agents, builders should also consider:
- **Cross-border and FX handling** – Many agentic transactions will be global; understanding exchange rates and messaging standards (e.g., ISO 20022\) is essential.
- **Security and key management** – Agents must safely store and use private keys or API credentials to authorize transactions.
- **Interoperability** – Bridging between fiat rails, stablecoins, and Layer 2 networks will be necessary for broad agentic functionality.
- **Regulatory compliance** – Even autonomous systems must comply with KYC, AML, and licensing frameworks.
These considerations go well beyond the capabilities and core competencies of most teams building agentic systems, and this is where the next two layers of the agentic commerce stack, infrastructure and banking, come into play. An AI bank will provide a trusted, regulated financial institution with the ability to address these considerations and the technical capabilities to empower teams working at the solution layer in the stack to build quickly and confidently, without having to worry about these layers of the stack
## **Looking Ahead**
In this post, we introduced payment rails as a foundational component of the A-Commerce protocol layer. They determine how money moves and settles—whether between banks, users, or agents.
In Part 2, we’ll explore the agentic protocols that sit atop these rails—enabling agent-to-agent and agent-to-merchant transactions with trust, programmability, and compliance built in.
# Building on Google's Agent Payment Protocol (AP2)
Published: 2025-09-18
URL: https://catena.com/blog/google-agent-payment-protocol-ap2
Summary: Google's new Agent Payments Protocol (AP2) is a big step for agent commerce. We're excited to build on this foundation by adding critical features like identity trust chains and reusable payment 'Receipts.'
---
Google just announced the [Agent Payments Protocol](https://ap2-protocol.org/) (AP2), and we couldn't be more excited. It's validating to see our partners at Google arrive at many of the same conclusions we have about the future of agent-driven commerce – some of which we document in the open source [Agent Commerce Kit](https://agentcommercekit.com) (ACK). We continue to build our own financial agents on top of this foundation every day.
Let's dive into AP2, what we like about it, and where we see opportunities to extend the foundation.
## A Solid Foundation
AP2 gets the fundamentals right: cryptographic signatures, verifiable credentials, and the need for trust infrastructure in agent-to-agent transactions. This validates several core principles we champion:
**Cryptographic identity is essential**: AP2 recognizes that agents need cryptographically verifiable identities to operate in commerce. This isn't only about preventing fraud, it's about creating accountability in autonomous systems.
**Verifiable Credentials fit agentic workflows**: Throughout their specification, AP2 emphasizes the use of Verifiable Credentials as the mechanism for establishing trust. This aligns with our conviction that signed digital credentials will become the backbone of agent commerce.
**Structured payment flows should be transparent, observable, and auditable**: AP2's mandate system (Intent Mandate → Cart Mandate → Payment Mandate) creates clear, auditable transaction flows that provide transparency and enable dispute resolution.
**Agentic payments should follow open interoperable standards**: We are committed to building solutions based on open standards rather than proprietary vendor-specific protocols. Agentic commerce should be accessible, extensible, avoid vendor lock-in, and benefit from broad industry collaboration.
## Evolving AP2
While we love AP2's direction, we see opportunities to build on it to provide additional features we need. For example:
### 1. Identity: Chains of Trust
AP2 introduces the concept of "manually curated allow lists of approved agents," but doesn't describe identity verification mechanisms.
Patterns such as [ACK-ID](https://www.agentcommercekit.com/ack-id/introduction) extend and augment the base concept with verifiable chains of trust which cryptographically tie agents to their accountable human or organizational owners. Rather than relying on static lists, this enables real-time identity verification with the legal compliance frameworks that enterprises need today.
### 2. Beyond One-Time Purchases: The Power of Receipts
As it stands today, AP2's mandates are transaction-specific and ephemeral—perfect for one-time shopping cart purchases. This is because they are designed for a single, auditable purchase flow, which is a great starting point. But agents can support richer commerce patterns by introducing a Verifiable Credential representing a “Receipt.”
A Receipt system enables persistent, reusable Verifiable Credentials that prove payment history. This unlocks:
- **Subscriptions**: Pay once, access repeatedly over a set timeframe or number of uses
- **Out-of-Band Payments**: Payments that happen outside the immediate browser session, enabling asynchronous agent commerce
- **Cryptographic Dispute Resolution**: When payment is charged but resources aren't delivered, clients have cryptographic proof for recourse
- **Cross-Platform Recognition**: Receipts work across different services and platforms
## How We Will Use AP2
AP2 validates our view that the industry benefits from cryptographic trust, verifiable credentials, and structured agent commerce protocols.
Here are a few ways we plan to incorporate AP2 into our work:
**AP2-Compatible Extensions to our frameworks, including ACK**: We are building support for AP2's mandate formats while extending them with identity trust chains, out-of-band payments, and Receipt capabilities.
**Identity Layer Integration**: AP2's "credential provider" role creates opportunities for specialization. ACK-ID serves as an identity layer that plugs into AP2's framework while providing the identity verification the ecosystem needs.
**Multi-Rail Payment Facilitation**: ACK's Payment Service architecture can serve as an AP2 and [x402](https://www.x402.org/)-compatible payment facilitator while providing the additional capability to route payments across multiple rails based on optimization criteria.
## The Big Picture
Agent commerce requires new infrastructure. Agents aren't just automated versions of human users—they operate at different scales, with different trust models, and with different economic patterns. Google's AP2 is a step forward for the space and helps establish de facto standards.
Together, we're building the foundational infrastructure for a new kind of economy: An economy where agents establish identity, operate within guardrails, prove capabilities, access resources, compensate participants, and create value and spread global prosperity in ways never before possible.
---
_Interested in building the future of agent commerce? Check out our [open-source documentation](https://www.agentcommercekit.com/overview/introduction)_
# ACK-Lab Developer Preview: Building Trust Infrastructure for Agentic Commerce
Published: 2025-09-03
URL: https://catena.com/blog/acklab-developer-preview
Summary: Learn how ACK-Lab solves the trust problem in AI agent commerce with cryptographic identity, secure wallets, and programmable rules. Get hands-on with developer tools that let autonomous agents negotiate, transact, and exchange value safely on testnet - from token swaps to data marketplaces. Build trustworthy agent-to-agent commerce in minutes.
---
**Note:** The ACK Lab developer preview has closed.
How do you give an AI agent money and not lose sleep over it?
We answered this question beginning with a foundational layer of protocols and patterns we identified in the open source [Agent Commerce Kit (ACK)](https://agentcommercekit.com). We shared a developer preview and demos of those patterns, ACK-Lab, for testing and experimentation.
## **From Protocols to Practice**
ACK defines open protocols and patterns built on internet standards for identity (ACK-ID) and payments (ACK-Pay), which are foundational to agentic commerce. These patterns illustrate how agents can reliably identify themselves and transact with each other.
Now, we're sharing a developer preview of ACK-Lab, which shows examples of how ACK works in practice. While anyone can implement ACK protocols from scratch, ACK-Lab provides a fast path to experimentation.
ACK-Lab gives your agents three capabilities:
1. **Identity**: Cryptographic credentials so agents can prove who they are and who controls them (based on ACK-ID)
2. **Wallet**: Secure testnet wallets so agents can transact safely with a proxy for stablecoins, testUSD, (based on ACK-Pay)
3. **Rules**: Policies that govern what agents are allowed to do
With the hosted ACK-Lab service and related demo apps, you can start testing and experimenting with how agents can use money in only a few minutes. ACK-Lab is based on the same infrastructure technology we’re developing for our AI-native bank.
## **See It In Action**
We are sharing two demos that showcase agent-to-agent commerce:
**Demo 1: Autonomous Token Swap**
Watch two agents negotiate and execute a USDC-to-SOL swap. One agent needs SOL, the other provides swap services. They:
- Exchange cryptographic credentials before any communication
- Fetch real-time prices from an oracle
- Generate payment tokens and process transactions
- Execute the swap with full audit trails
The entire flow happens autonomously, with ACK-Lab rules controlling maximum transaction sizes, minimum balances, and spending limits.
Explore the Swap Demo [code on Github](https://github.com/catena-labs/ack-lab-demo-swap).
### **Demo 2: Data Marketplace Negotiation**
See agents negotiate access to premium datasets. A researcher agent discovers data from a provider agent, then they:
- Verify each other's identities before negotiating
- Dynamically negotiate prices within policy bounds
- Process payment through ACK-Pay
- Grant time-limited access tokens
The negotiation isn't hardcoded. Agents find mutually acceptable prices based on the rules you set.
Explore the Marketplace Demo [code on Github](https://github.com/catena-labs/ack-lab-demo-data-marketplace).
## **Getting Started in Minutes**
The setup flow is designed to get you experimenting quickly:
**Step 1: Register Your Agents** → The developer preview is now closed.
**Step 2: Run a Demo** → Follow the instructions to clone and configure the demo.
**Step 3: Experiment and Iterate** → This is where it gets interesting. Run a demo and watch a transaction get blocked by a spending limit. Switch back to ACK-Lab, adjust the policy, and run it again. See how different rules change agent behavior. Lower the daily limit and watch negotiations fail. Require stricter verification and see trusted transactions succeed.
**Step 4: Build Your Own** → Once you’ve gotten a handle on how the system works, build your own agents and experiment with how they can use identity and stablecoins to send and receive payments.
The entire loop, from agent creation to policy adjustment, takes about 10 minutes.
## **Built for Builders**
ACK-Lab runs on testnet, so you're using test dollars — all the complexity of real transactions without the risk. The ACK-Lab SDK handles credential exchange, payment processing, and policy enforcement. With these capabilities, you can focus on what your agents should do, not how to make them trustworthy.
ACK-Lab was a hosted service, but the underlying ACK protocols remain open source. You can implement them directly in your own infrastructure.
## **Join the Experiment**
We need your feedback. What policies do your agents need? What's too complicated? What's missing?
- Join our [Discord](https://discord.gg/catenacommunity) to share what you're building
- Explore the ACK protocol documentation at [agentcommercekit.com](http://agentcommercekit.com)
- Tell us what agent commerce infrastructure should look like
# The Agentic Finance Stack: Building the Financial Capabilities for AI Agents
Published: 2025-08-13
URL: https://catena.com/blog/the-a-commerce-stack-building-the-financial-capabilities-for-ai-agents
Summary: The agentic finance stack is a six-layer framework designed to enable AI agents to act as primary economic actors in B2B and B2C commerce. This stack addresses the limitations of current financial systems, which are not equipped for the speed and scale of AI transactions, by introducing new protocols, infrastructure, and re-imagined financial services.
---
Agentic commerce and finance is the next evolution of **B2B and B2C commerce**. It’s an era where AI agents, acting on behalf of businesses and consumers, become primary economic actors. These agents will manage supply chains, procure goods, pay invoices, and handle complex financial tasks with unprecedented speed and scale.
But to unlock this potential, we can't simply plug these agents into our existing financial system. AI agents have dramatically different capabilities and limitations from human actors. They are incredibly fast and can execute millions of transactions without fatigue, but they lack human intuition and are vulnerable to novel forms of fraud and manipulation. A single misconfigured agent could create significant financial damage in seconds.
## The New Agentic Finance Stack
Realizing the promise of AI economy requires a new technology and financial services stack, one designed from the ground up to provide the necessary guardrails, security, and services for AI agents to conduct commerce safely and effectively. This stack consists of six layers.
### 1. Protocols and Payment Rails
At the base of the stack are the protocols for communication and value exchange. This layer must be machine-native. While it will incorporate traditional payment rails like ACH and card networks, it will heavily rely on new, more programmable infrastructure.
**Stablecoins** will be critical, offering 24/7 settlement, low transaction costs, and programmability. This layer includes new identity and payment protocols such as Agent Commerce Kit (ACK) and x402. It also includes new protocols for how agents interact with each other and with external tools, such as Google’s Agent2Agent Protocol (A2A) and Anthropic’s Model Context Protocol (MCP). This layer will incorporate standards such as decentralized identifiers (DIDs) and verifiable credentials (VCs) from the W3C. These protocols will make it possible for there to be an open, interconnected system to establish identity and trust and to execute payment flows without confining agents to closed networks.
### 2. Infrastructure
Sitting on top of the protocols is the technical infrastructure—the non-regulated toolkits, frameworks, and systems that make the protocols usable. This is where developers work. It includes:
- **Account and Wallet Infrastructure:** Secure systems for agents to hold, manage, and deploy digital assets.
- **Cross-Chain Transfer:** Systems for optimizing and managing stablecoin movement across different chains and conversions between multiple stablecoins.
- **APIs and SDKs:** Tools that allow developers to integrate their agents into the financial system.
- **Agent Identification:** A crucial capability for a-commerce. This is the equivalent of a digital passport for AI agents, allowing them to identify themselves and establish trust with other agents and systems.
- **Rules Enforcement:** This is where programmable guardrails are technically enforced. For example, a developer can set rules directly in the infrastructure to limit an agent's spending to $500 per transaction, restrict it to pre-approved vendors, or require human approval for payments over a certain threshold.
- **Payment Rail Infrastructure:** These are the technologies to easily integrate and use traditional payment rails such as cards, ACH, and real-time payments (RTP).
This layer provides the fundamental building blocks for secure agent operations before any regulated activity takes place.
### 3. Regulated Banking and Financial Services
This is the layer of trust and compliance, where financial services are provided by regulated entities like banks and Money Services Businesses (MSBs). These services must be re-imagined for an agent-centric world.
- **KYC/KYB becomes KYA:** "Know Your Customer" and "Know Your Business" protocols must expand to include "Know Your Agent." Financial institutions will need processes to verify, monitor, and understand the behavior of the AI agents operating on behalf of their clients.
- **Advanced Risk and Fraud Management:** Anti-Money Laundering (AML) and Counter-Terrorist Financing (CTF) systems need to be trained to detect anomalous patterns specific to AI agents, which can differ wildly from human patterns.
- **Human-in-the-Loop Controls:** This layer ensures that while agents can operate autonomously, there are always clear lines of human accountability and intervention. This is a non-negotiable requirement for managing risk.
- **Deposit Taking and Payment Flows:** Taking deposits and handling pay-in, pay-out, and conversion between stablecoins and fiat currencies.
- **Treasury Management:** Providing treasury services to generate yield on stablecoin and fiat deposits, including float in payment flows.
- **Credit:** Accessing loans and other credit facilities.
- **Auditability and Observability:** Actions and transactions executed by AI workflows must be observable and reportable in a compliant, privacy-preserving and human-aligned manner.
Services like deposit taking, card issuing, and custodial stablecoin accounts will all be adapted to serve agents as the primary account users.
### 4. Fintech Services
Building on the regulated banking core, the fintech services layer packages these complex capabilities into more easily consumable products. This includes Banking as a Service (BaaS) providers who offer agent-ready bank accounts via an API, Payment Service Providers (PSPs) that specialize in orchestrating payments for agentic finance platforms, and embedded finance solutions that allow non-financial companies to integrate these new capabilities directly into their products. This layer acts as an accelerator, allowing businesses to leverage agents in financial transactions without needing to build a direct, deep integration with a bank.
### 5. Vertical and Horizontal Business Solutions
This is where the technology meets specific, real-world business needs. Here, the underlying stack is used to build proprietary corporate systems or commercial software products.
- **Horizontal Solutions:** These address common business functions, such as agent-driven accounts payable (AP) and receivable (AR) management, intelligent payroll, and automated international remittances.
- **Vertical Solutions:** These are industry-specific applications. For example, a logistics company might use an agent-based system for supply chain management that automatically orders materials, tracks shipments, and pays suppliers upon delivery confirmation.
- **Consumer Solutions:** These might include agentic services for shopping, organizing travel, or investment and budget management.
- **Agent Solutions:** Services that give agents access to commercial capabilities such as proprietary data sets, analytical tools, or tools for executing real-world actions.
This is the layer where most businesses will directly experience the benefits of agentic business transactions through increased efficiency and automation.
### 6. End Users
The final layer is composed of the businesses and consumers who use the solutions. A manufacturing company's procurement department uses a vertical solution to reduce costs, or a small business owner uses a horizontal accounting platform to automate invoicing and payments. For consumers, this could mean a personal finance agent that negotiates bills, finds the best insurance rates, and executes trades based on pre-defined goals. The complexity of the underlying stack is abstracted away, and users simply benefit from the outcome: smarter, faster, and more efficient commerce.
## The Path Forward
Building this stack will be an evolutionary process. It will not be a "rip and replace" of the existing financial system. Instead, it will be a collaborative effort. **Existing players** --- banks, payment networks, and established fintechs—will move to adapt their offerings for the agent economy. They bring regulatory expertise, trust, and scale. At the same time, **new entrants** will emerge, driving disruptive innovation at every layer in the stack.
As recognized by analysts at firms like Gartner, who have highlighted the importance of AI Trust, Risk, and Security Management (AI TRiSM), the focus must be on creating a robust and secure environment for AI to operate. The success of AI economy and agentic commerce hinges on our ability to build a financial system where AI agents can act as powerful economic engines while being safely tethered to human oversight and sound risk management principles. This six-layer stack provides the pragmatic roadmap to get there.
## Where We'sre Building
At Catena Labs, we’re working on advancing the work at several layers in this stack. We’re developing protocols and pattern languages like the open source Agent Commerce Kit (ACK) and contributing to other protocols we believe will be critical at the protocol layer. We’re also building infrastructure that will be required for developers integrating and using these lower-level protocols by enabling easy adoption of agent identity, stablecoin wallets, and rules. Our main focus is working at the banking and financial services layer to provide the regulated, reliable, and trusted financial services that will be required for AI agents to participate in the economy.
# A New Addition to our Team: Charting the Future of Policy and Regulations for Agentic Commerce
Published: 2025-07-29
URL: https://catena.com/blog/new-addition-to-our-team-charting-the-future-of-policy-and-regulations-for-agentic-commerce
Summary: Catena announced that Sharda Caro Del Castillo is joining as Chief Legal & Business Officer to help navigate the regulatory landscape for AI-driven financial services. She brings 25 years of experience from companies like PayPal and Square, and will focus on partnering with regulators to adapt the financial system to agentic commerce, AI agents, and stablecoins. Caro el Castillo is motivated by making financial ecosystems more accessible and inclusive and believes AI will bring greater efficiency and equity to financial products.
---
Today, we’re announcing that [Sharda Caro Del Castillo](https://www.linkedin.com/in/sharda-caro-del-castillo-796b28/) is joining Catena as the Chief Legal & Business Officer. Why you should care takes a bit of an explanation.
We’ve written in other posts about how we’re building a regulated financial institution for AI agents. The challenge is that the regulators and policy-makers who oversee the existing financial system have only barely begun to make sense of how AI agents and stablecoins will transform financial services globally.
We believe the work regulators do is important. Financial systems need well-designed guardrails to protect the public interest and create a level playing field. Unfortunately, poorly designed regulatory schemes can have the opposite effect – stifling innovation, limiting prosperity, and creating regulatory capture for incumbents.
As a result, we’re taking a proactive approach to partnering with regulators and policy makers to ensure that the potential for agentic commerce, the value of AI agents, and the power of stablecoins isn’t capped by a failure to adapt the regulatory system to a dramatically changing technological and economic landscape.
That’s why we’ve hired an exceptional Chief Legal and Business Officer. Sharda has spent the last 25 years guiding some of the most influential fintech companies through legal complexity and growth. Check out her LinkedIn profile for the full rundown. Given how important this work is to the developers and customers we work with, we’re sharing a Q&A with Sharda to introduce her perspective and what she’ll be focused on at Catena.
**What brought you to Catena Labs, and how does this opportunity feel different from your previous roles?**
I have spent the last few years as a board member. I have loved keeping up with the market and understanding governance from a new perspective. As I started to think about being an operator again, I knew that I wanted to build with a broad mandate, which to me meant being part of an early-stage company. When I was first introduced to Sean, everything seemed to fall together in a way that I could not have orchestrated. A known founder in the early stages of building a groundbreaking product that has the opportunity to change how we move money globally; it was the perfect fit.
**You’ve led legal teams at some of the most innovative companies in fintech and payments–PayPal, Square, Airbnb, Affirm–what originally drew you to this space, and what continues to motivate you?**
My journey into fintech was the byproduct of both timing (the passage of the Dodd-Frank Act in 2010) and attitude (my willingness and confidence that change breeds opportunity). In a way, I was a bit of a “non-lawyer” lawyer during that time, choosing not to look solely at risk mitigation and management, and instead to focus on how we could rethink systems and products to serve customers better. As someone who grew up in developing countries, I continue to be motivated by working on products that have the potential to change our financial ecosystem to be more accessible and inclusive on a global scale.
**You’ve worked on groundbreaking products like 'Pay in Four' at Affirm. How do you see AI transforming how financial products are designed, delivered, and regulated?**
I think that we’re at the cusp of a new era in terms of product design, and even delivery, when we think about AI agents providing services. In the case of financial products, I envision greater efficiency, equity, and accuracy as digital agents employ best practices to deliver consistent and compliant products and services.
**You’ve navigated complex regulatory environments across legacy banks, fintech disruptors, and global platforms. How does AI-driven finance compare?**
I am excited to learn about the opportunities presented by AI-driven finance. There are some obvious efficiencies to be gained by introducing digital agents into the financial ecosystem, which would be immensely beneficial. But, I also think that AI may help us expedite and move exponentially to improve some of the consumer protection issues that regulators worry about; this is an area that I look forward to exploring with Sean and the team.
**Catena’s mission to build an AI-native financial institution is ambitious. What excites you most about this vision, and where do you see the greatest legal and regulatory challenges?**
The fact that Catena has an ambitious mission is the draw. I like the challenge of big problems, and in particular, working on novel solutions to challenges that have yet to be solved. In my experience, I’ve found that technology generally improves the financial services experience. In the case of an AI-native financial institution, we’ll need to think through legal and regulatory issues as applied to digital agents; this will require communication, clarification, and collaboration with various industry agencies, regulators, and commercial partners. However, I’m confident that those conversations have the potential to shift how we interact with money at a societal level.
**The rise of agentic commerce introduces new questions around identity, authentication, and liability. How do you see the legal framework evolving to keep pace with autonomous systems?**
The law is always evolving to keep pace with innovation. I remember when prepaid cards were in their nascent stages of development, and regulators were concerned with BSA/AML risk. At that time, I advocated for ways in which safety and soundness could be balanced with serving the unbanked and underbanked. I see the current legal challenges in the same light—we will have to build trust and clear paths of communication with our regulators while we innovate so that the intent and impacts are well understood, even if the laws are still catching up.
**As someone who has helped shape the legal strategy of household-name companies, how do you balance risk mitigation with speed and experimentation in emerging tech?**
This is really an art more than a science. We don’t always know what is around the bend until we turn the corner. But if we start to put too many safeguards in place before one can even get to the corner, we thwart progress. So, risk management during the nascent stages of development needs to be thought of in the context of meeting public policy goals while moving quickly to experiment and understand customer experience. The legal strategy at this stage in Catena’s lifecycle is to ensure that product development is consistent with the public interest; this tells us that there’s a path to compliance.
# Why AI Builders Should Celebrate Passage of the GENIUS Act
Published: 2025-07-18
URL: https://catena.com/blog/why-ai-builders-should-celebrate-passage-of-the-genius-act
Summary: Stablecoins are the only money that makes agentic commerce possible at scale. Traditional payments create friction everywhere—days to clear, high fixed fees that kill micropayments, geographic barriers, and human authorization requirements that defeat autonomy. Stablecoins solve this with instant 24/7 settlement, sub-penny costs, global accessibility, and programmability. This unlocks agent-to-agent commerce where specialized AI agents can pay each other for services in real-time.
---
The passage of the GENIUS Act marks a historic milestone for AI builders.
In case you're new to the stablecoin party, stablecoins are a form of cryptocurrency that is pegged to the value of the dollar (or another fiat currency) because it is backed by a 1:1 reserve of cash or treasury bills. Stablecoins combine the reliability and low volatility of government-backed fiat currency with the functionality of public blockchains.
Stablecoins are the ideal form of digital money for AI agents. They offer 24/7 near-instant settlement, very low transaction costs, the ability to use dollars worldwide, and programmability. Stablecoins make it much easier for AI agents to conduct commerce. They provide a way for agents to hold balances, make payments, and receive payments for a broad array of use cases.
The barrier to mainstream stablecoin adoption has been the lack of a clear regulatory framework and legal standing for US dollar stablecoins. Without regulatory clarity from the federal government, there was uncertainty about the legal and compliance risks of using stablecoins. This regulatory uncertainty held back adoption by established financial service providers such as banks, payment networks, and fintechs.
The passage of the GENIUS Act creates a clear legal status and regulatory scheme for stablecoins. This clarity will open the floodgates. Now, AI builders can be confident that stablecoins from regulated issuers, such as USDC from Circle (CRCL: NYSE), will be reliable, trustworthy, and overseen by federal and state regulators. In the coming months, you’ll see rapid adoption of stablecoins for mainstream payments use cases, including agentic commerce.
Regulated stablecoins and agentic financial solutions will have broad positive impacts on people and businesses, for example:
- Faster, less expensive global remittances
- New ways to discover and buy products online
- Opportunities for small businesses to reach new customers
- Better supply chain payment and financing
- Global treasury management solutions
- New ways to work with vendors globally
At Catena, we’re building on the opportunities created by the GENIUS Act. By providing a regulated financial institution that is purpose-built for the needs of AI agents, we’ll give your agents banking, payment, and treasury products that they can use to engage in global commerce safely and securely.
Welcome to the new era of AI money.
# Why Stablecoins Are the Perfect Money for AI Agents
Published: 2025-06-26
URL: https://catena.com/blog/why-stablecoins-are-the-perfect-money-for-ai-agents
Summary: Stablecoins are the only money that makes agentic commerce possible at scale. Traditional payments create friction everywhere—days to clear, high fixed fees that kill micropayments, geographic barriers, and human authorization requirements that defeat autonomy. Stablecoins solve this with instant 24/7 settlement, sub-penny costs, global accessibility, and programmability. This unlocks agent-to-agent commerce where specialized AI agents can pay each other for services in real-time.
---
We've covered how [stablecoins have reached escape velocity](https://catena.com/blog/stablecoins-meet-ai-perfect-timing-for-agent-commerce) as mainstream financial infrastructure. Now let's talk about why this matters specifically for the future we're building at Catena Labs.
AI agents are becoming independent economic actors, but they need financial infrastructure designed for their capabilities. Traditional payment systems create friction at every step—from authorization delays to geographic restrictions to cost structures that make micropayments impossible. Stablecoins aren't just better payment rails for agents; they're the only money that makes agentic commerce possible at scale.
The infrastructure is finally ready. Now we need to build the banking layer that ensures agentic commerce operates safely and compliantly.
## Why Traditional Money Breaks Down for AI Agents
AI agents operate fundamentally differently from humans, and traditional financial systems weren't built for their needs.
**Speed mismatches create bottlenecks.** Traditional payments take days to clear while agents operate 24/7. A research agent that identifies valuable market data on a weekend sits idle until banking hours resume. Supply chain agents that spot arbitrage opportunities miss them waiting for international wires to process.
**Cost structures make micropayments impossible.** Fixed fees of $15-50 per wire transfer mean agents can't make granular transactions. An agent that could optimize cloud costs by switching providers multiple times per day is blocked by transaction costs that exceed the savings.
**Geographic barriers limit global operations.** Agents work across jurisdictions seamlessly, but traditional banking creates friction at every border. Correspondent banking delays, FX spreads, and compliance requirements in each jurisdiction make global agent operations prohibitively complex.
**Human-centric authorization blocks autonomy.** Every transaction requires human approval somewhere in the chain, defeating the purpose of autonomous operations. Treasury agents that could optimize cash management in real-time instead wait for manual approvals that take hours or days.
For a deeper dive into these limitations, check out our post "[Beyond Automation: Why AI Agents Need Their Own Money](https://catena.com/blog/beyond-automation-why-ai-agents-need-their-own-money)."
## Stablecoins: Money Built for Machines
Stablecoins solve these problems because they were designed for the speed, scale, and costs required for agent-to-agent transactions.
**Instant settlement enables real-time operations.** Stablecoins settle 24/7 in seconds, not days. Agents can execute transactions when opportunities arise, not when banks open. A treasury agent can rebalance portfolios based on overnight market movements. A supply chain agent can secure better rates the moment they become available.
**Near-zero costs unlock micropayments.** Transaction costs under a penny make granular payments economically viable. Agents can pay per API call, per computation, or per insight. A research agent can purchase individual data points rather than expensive dataset subscriptions. Multiple specialized agents can collaborate on complex tasks, each compensated for their specific contribution.
**Global by default eliminates geographic friction.** No correspondent banks, no FX delays, no jurisdiction-specific requirements for basic transfers. An agent managing global operations can move funds between regions instantly. Cross-border B2B payments that traditionally take days and cost hundreds in fees happen instantly for pennies.
**Programmability creates new possibilities.** Stablecoins support smart contract logic and automated conditions. Future applications could include escrow arrangements that release payments based on verified outcomes or automated recurring payments that adjust based on usage patterns.
## The Developer Value Proposition
For AI engineers building agentic systems, stablecoins offer compelling advantages over traditional payment integration.
**Lower integration costs.** No need to navigate complex banking partnerships or compliance requirements for basic payments. Simple APIs handle wallet creation, balance management, and transfers. A developer can integrate payments in hours, not months.
**Significantly reduced fees.** Save 70-90% compared to traditional payment processors. A system processing thousands of small transactions per day can save substantial costs. More importantly, the economics make previously impossible use cases viable.
**Faster development cycles.** Test payments work the same as production payments. No separate sandbox environments or lengthy approval processes. Deploy globally without a jurisdiction-specific banking setup.
**Agent-to-agent commerce becomes practical.** Direct value exchange between specialized agents creates new business models. A translation agent can pay a verification agent for quality checks. A market analysis agent can compensate multiple data providers based on accuracy. These collaborative workflows generate value impossible with traditional payment systems.
Our goal is to build simple APIs that handle complex regulatory requirements behind the scenes. You get the benefits of stablecoin payments—speed, low cost, global reach—with the compliance infrastructure that makes enterprise adoption possible.
## The Catena Labs Vision
At Catena, we're building the first AI-native financial institution on the proven stablecoin foundation. Our approach is regulatory-first: becoming a licensed financial institution rather than finding fintech workarounds.
The end state is AI agents as full economic participants with proper oversight and agents paying other agents for specialized services. Autonomous treasury management that optimizes cash positions in real-time. Global operations that aren't constrained by traditional banking limitations.
Stablecoins provide the foundation layer that makes this possible. But safe, compliant, agentic commerce at scale requires purpose-built banking infrastructure. That's what we're building at Catena Labs—the missing piece that turns stablecoin potential into agentic commerce reality.
The rails are ready. The regulatory framework is emerging. Now it's time to build the banking layer that makes agentic commerce safe, compliant, and accessible to every AI engineer building the future.
# The Critical Need for Verifiable AI Agent Identity
Published: 2025-06-18
URL: https://catena.com/blog/the-critical-need-for-verifiable-ai-agent-idenity
Summary: AI agents face critical identity verification challenges when conducting autonomous transactions, as they cannot rely on human trust systems like licenses or credentials to establish authenticity and accountability. The Agent Commerce Kit (ACK) framework addresses this gap using decentralized identifiers (DIDs) and verifiable credentials (VCs) to enable secure agent-to-agent commerce, verified marketplaces, and trusted autonomous transactions.
---
Imagine you’ve built a personal shopping AI agent, and it’s ready to make a purchase from the Amazon sales agent. Before any money can be exchanged, the agents face critical identity problems: How does your agent know it’s interacting with the real Amazon agent? How does the Amazon agent know that your agent is who it claims to be? How can both agents confirm the legal entities that are liable if something goes wrong with the transaction? How can your agent check if there are any rules regarding transactions with the Amazon agent, without a consistent and reliable agent identity to reference?
These identity problems appear again and again in a wide range of circumstances where independent agents could collaborate with each other or engage in commerce.
# The Agent Identity Gap
Long ago, people learned that identity and trust are critical for commerce and collaboration, so humans have a wide range of mechanisms to prove their identities in ways that are appropriate in the context of a transaction. The bouncer lets you into the bar because he trusts the state has issued a driver's license with a valid birthday. You trust a doctor because she has credentials to prove her competence and other reputation signals, such as the hospital where she works. You trust your bank with your savings because it can prove that it’s chartered. Your bank trusts you with a credit card after they’ve collected several proofs of your identity and checked your creditworthiness.
These human systems of proving identity and establishing trust do not make sense for AI agents. Agents operate 24/7 across borders. They make decisions in milliseconds. They can be cloned, modified, or compromised in ways humans cannot. Identity and trust systems built for humans won't work for AI agents.
The lack of agentic identity creates a significant gap between what we can envision agents accomplishing and what they can actually do. It limits the potential for agentic collaboration and commerce in many ways. For example:
- **Payments**: Executing agent-to-agent or agent-to-service payments has serious risks for fraud and illegal activity, which means that without identity, agents can’t participate in commerce. Moreover, because agents are not legal persons, a critical part of agent identity is reliably linking them to businesses and individuals, creating a chain of ownership.
- **Security**: Without verifiable identity, an agent could make arbitrary claims about capabilities or ownership in an agent registry, and then, when it takes on a task, act maliciously, intercepting sensitive data, misdirecting actions, or sabotaging entire workflows. (Security researcher Tom Neaves [demonstrated how malicious agents can game multi-agent systems](https://www.trustwave.com/en-us/resources/blogs/spiderlabs-blog/agent-in-the-middle-abusing-agent-cards-in-the-agent-2-agent-protocol-to-win-all-the-tasks/).)
- **Quality** \- Even if an agent isn’t malicious, it may simply be bad at the job it claims to do. Without a system to reliably establish and maintain reputations for agents, an agent can’t know if the counterparty agent it’s working with is really the best choice for the job.
Today, there are methods for authentication that agents could employ, such as API keys and bearer tokens. However, these methods can only confirm the possession of a valid credential. They can’t verify that the agent using the token is actually the agent it claims to be, because the credentials could be stolen or the capabilities could be misrepresented. Moreover, there is a range of other claims agents will need to make, so a more flexible system for identity is required.
# Closing the Gap
In the [Agent Commerce Kit (ACK)](https://www.agentcommercekit.com/overview/introduction) framework, we’ve proposed a pattern ([ACK-ID](https://www.agentcommercekit.com/ack-id/introduction)) for agent identity that uses open standards to address the core identity challenges. ACK-ID works with and builds on other protocols that contribute to closing the identity gap.
## Defining an Agent
The first step to establishing an agent identity is to describe the agent. We expect there will be several broadly adopted approaches to this. For example, the [Google Agent2Agent (A2A)](https://google-a2a.github.io/A2A/latest/) protocol addresses this problem with an “[Agent Card](https://google-a2a.github.io/A2A/latest/specification/#5-agent-discovery-the-agent-card),” which is a specific format and protocol for describing an agent and its capabilities in a machine-readable format. However, while a description is necessary, it is not sufficient.
## Establishing Identity
The next layer is establishing identity. ACK-ID uses the [decentralized identifier (DID) standard](https://www.agentcommercekit.com/ack-id/standards#decentralized-identifiers-dids) from the W3C. Think of DIDs as globally unique, permanent digital IDs that agents (and their owners) can create and control without needing permission from a central registration authority like ICANN for domain names or like a government for national IDs. Unlike traditional usernames or identifiers issued by a single company, DIDs are independent and cryptographically secured.
DIDs offer a number of advantages as IDs for agents:
- Already a well-established W3C standard
- Flexible, easy to adopt, and scalable
- No dependency on a centralized authority
- Resilient to fraud and attacks through the use of public key cryptography
- Easily resolved by other agents
- Convenient for conveying other information, such as access endpoints
ACK-ID uses DIDs to identify both agents and their owners.
## Verifying Claims
Having a clear description of an agent, such as an A2A Agent Card, and a way to identify the agent using a DID are necessary building blocks, but not sufficient. The final component is a mechanism to easily and confidently verify claims an agent makes.
While DIDs provide unique identifiers, [Verifiable Credentials (VCs)](https://www.agentcommercekit.com/ack-id/standards#verifiable-credentials-vcs) provide a standard way to make verifiable statements about those identifiers. Think of VCs as digital, tamper-proof versions of physical credentials like a driver’s license, a diploma, or a membership card, but designed for the digital realm.
VCs contain statements or “claims” made by a trusted Issuer about a Subject (e.g., an agent identified by its DID). These claims are digitally signed by the Issuer using their own DID and cryptographic keys. This signature allows anyone (as a Verifier) who trusts the Issuer to verify the authenticity and integrity of the claims about the Subject.
For payments, the most important claim an agent needs to prove is the identity of the entity that controls them. Without knowing the individuals or businesses that are controlling the agents in a transaction, the agents can’t be confident that they are sending or receiving payments in a way that is safe and legal.
Beyond ownership, it’s easy to imagine that agents will want to verify other claims made by agents. For example, claims about capabilities and functionality. Moreover, agents will need mechanisms to evaluate the reputations of other agents. DIDs, VCs, and the ACK-ID pattern language provide the building blocks that can be extended for these different use cases.
# Unlocking Value
A robust system for verifiable agent identity opens up a wide range of use cases. Here are a few examples:
- **Secure Agent-to-Agent Payments**: When your research agent purchases a dataset from a data provider, identity verification ensures the payment goes to the right place. The data provider proves it belongs to "DataCorp Inc" before your agent releases payment. If the data never arrives or arrives corrupted, you know exactly who to pursue for resolution. This accountability makes agents willing to transact with new partners.
- **Agent Registries and Discovery**: Verified agent marketplaces are emerging where developers register their agents with proof of identity. Just as app stores verify developer accounts before allowing apps, these registries verify agent creators. Your coding assistant can search for debugging tools, knowing each result links to a real developer who can be held accountable. No more anonymous agents claiming capabilities they don't have.
- **Preventing Service Abuse**: Customer service agents face constant attacks from bad actors trying to extract sensitive information. With identity verification, your support agent can verify that incoming requests come from legitimate user agents, not sophisticated attack tools. Similarly, when your agent needs human verification for a complex task, it can confirm the human reviewer's identity before sharing confidential data.
- **Specialized Service Networks**: As agents specialize, they form ecosystems where a writing agent might pay for fact-checking services, a coding agent might purchase security audits, or a design agent might buy stock photography licenses. Each transaction requires trust. Identity verification enables these micro-economies to flourish without manual oversight.
As fintech analyst Simon Taylor notes in "[The checkout page is dead](https://www.fintechbrainfood.com/p/the-checkout-page-is-dead)", we're witnessing a fundamental shift where "commerce is moving upstream to the chatbot." When agents handle payments autonomously at the "point of intent," identity becomes the only trust anchor left.
These aren't theoretical scenarios. Major tech companies and financial institutions are deploying these systems today. Visa and Mastercard have announced agent identity as a top priority. Google, OpenAI, and others are building agent authentication into their platforms. The infrastructure for trusted agent commerce is being built right now.
# Getting Started
If you’re ready to dig into agent identity more, take a look at the documentation and demos at the [ACK website](https://www.agentcommercekit.com/overview/introduction).
# Stablecoins Meet AI: Perfect Timing for Agent Commerce
Published: 2025-06-10
URL: https://catena.com/blog/stablecoins-meet-ai-perfect-timing-for-agent-commerce
Summary: Stablecoins hit escape velocity—$15.6 trillion processed, matching Visa's volume. With Circle's IPO and major payment companies racing to build infrastructure, digital dollars are becoming essential financial rails. Perfect timing: AI agents need instant, programmable money to operate autonomously. Here's how stablecoins are solving the infrastructure problem holding back agentic commerce.
---
[Circle's IPO last week](https://catena.com/blog/how-the-circle-ipo-accelerates-agentic-commerce) wasn't just another crypto milestone—it was the moment stablecoins officially crossed the chasm from crypto experiment to essential infrastructure. We're watching a fundamental shift happen in real-time, with stablecoins processing $15.6 trillion in 2024 (matching Visa's volume) while every payment company scrambles to build stablecoin capabilities.
This isn't gradual adoption anymore. It's escape velocity—the point where momentum becomes unstoppable. And the timing couldn't be better for AI agents entering the economy as autonomous actors. We're excited because stablecoins are finally providing the infrastructure that agentic commerce has been waiting for.
## Three Forces Converging
**Regulatory uncertainty is clearing up.** The GENIUS Act has bipartisan backing in Congress, Circle successfully went public, and accounting standards now allow stablecoins as cash equivalents on balance sheets. While not complete, the regulatory picture is becoming clearer for institutional adoption.
**Corporate adoption is accelerating.** Major payment companies are moving beyond pilots to production infrastructure \- not gradual testing, but serious buildout across multiple fronts simultaneously.
**Business efficiency is driving adoption.** Companies need faster, cheaper payment rails for modern operations. Traditional systems with multi-day settlement and high fees create real friction, especially for businesses operating globally or serving digital-first customers.
As Simon Taylor puts it: "[Every fintech company will be a stablecoin company](https://www.fintechbrainfood.com/p/stablecoins-are-a-new-platform)." The infrastructure comparison is apt—this feels like internet protocols in the 90s. Once critical mass hits, adoption becomes inevitable because the alternatives can't compete.
## The Infrastructure Land Grab
The timeline tells the story. Stripe announced [stablecoin financial accounts](https://stripe.com/newsroom/news/sessions-2025#:~:text=Reinventing%20global%20money%20management%20with,acquisition%20of%20stablecoin%20platform%20Bridge.) in 101 countries while Bridge launched USDB. Visa partnered with Bridge on stablecoin-linked cards that let users spend from stablecoin balances at 150M+ merchant locations, starting in Latin America. Mastercard unveiled end-to-end stablecoin capabilities including merchant settlement options and on-chain remittances. Coinbase created a [new payment standard](https://www.coinbase.com/en-sg/developer-platform/discover/launches/x402) designed for AI agents.
These companies are positioning for market share before regulation fully settles and before competitors catch up. The winners are building compliant, robust infrastructure while the window is open.
## Real Business Value
Previous crypto waves were driven by speculation. This wave is driven by businesses solving actual problems.
The numbers are stark. International remittances average 6.62% in fees—a transfer of $200 costs $13.24. With stablecoins, that same transfer costs less than $0.01. B2B payments take 3-7 days and cost $14-150 per $1,000 transacted, passing through up to five intermediaries that each take a cut. Stablecoins settle instantly for near-zero fees.
For brick-and-mortar businesses, the impact is immediate. Coffee shops lose 15% of every $2 transaction to credit card fees—only $1.70-$1.80 actually reaches the shop. According to [a16z's analysis](https://a16zcrypto.com/posts/article/stablecoin-guide-what-why-how/), Walmart could increase profitability by over 60% by eliminating \~$10B in annual payment fees, while Kroger could potentially double profits given sub-2% margins.
Network effects are accelerating. More users drive merchant adoption, which drives infrastructure investment, which drives more users. Unlike earlier payment network attempts, stablecoins operate on open, global infrastructure that doesn't require proprietary merchant adoption.
## The AI Commerce Connection
This timing isn't coincidental. Stablecoin infrastructure maturity is converging with AI agents becoming autonomous economic actors.
AI agents increasingly need to execute financial transactions independently. Research agents that could optimize spending if they could pay for datasets directly. Supply chain agents that identify cost savings but need human approval for every transaction. Personal finance agents that spot better deals but can't act on them.
Traditional banking wasn't built for agentic commerce, but stablecoins were. Digital money that settles instantly, operates 24/7, and costs almost nothing to transfer—exactly what autonomous agents need.
While stablecoins solve the "rails" problem, AI agents still need banking-layer services: identity verification, compliance oversight, treasury management, and human-in-the-loop controls. That's the infrastructure layer we're building at Catena Labs on top of the proven stablecoin foundation.
## What's Next
Mainstream adoption is accelerating beyond cross-border payments. Stablecoins are becoming the default for B2B settlements, treasury management, and any application requiring instant, flexible transactions.
The regulatory framework is solidifying with the GENIUS Act. AI agents are becoming sophisticated enough to be primary financial services users. The infrastructure evolution is moving from basic payment rails to complete financial ecosystems.
Stablecoins have reached escape velocity. The only question is how quickly the rest of the financial system adapts to this new reality.
# How the Circle IPO Accelerates Agentic Commerce
Published: 2025-06-05
URL: https://catena.com/blog/how-the-circle-ipo-accelerates-agentic-commerce
Summary: Circle's historic IPO validates stablecoins as essential financial infrastructure, creating the foundation needed for agentic commerce. We explore how USDC's mainstream adoption directly enables AI agents to participate in the economy through programmable money that settles instantly, operates 24/7, and costs almost nothing to transfer - capabilities traditional banking can't match for autonomous commerce.
---
It’s a big day for everyone who's been building the future of digital money and AI because Circle, the creator of the USDC stablecoin, completed its initial public offering on the NYSE. Congratulations to the entire Circle team on this incredible achievement!
Many people on our team played roles in Circle’s growth over the years, including our CEO, Sean, who co-founded Circle, so we’re psyched to see our colleagues and friends reach this historic milestone. We’re also excited about the momentum the IPO creates for the adoption of USDC as foundational financial infrastructure for the global economy. USDC will play a critical role in agentic commerce.
## From Stablecoins to Agentic Commerce
Circle and USDC are leading the movement of stablecoins into the mainstream. Major financial institutions are becoming comfortable with the value and features that stablecoins unlock. Regulatory clarity is emerging through legislation like the GENIUS Act in the U.S. and MICA in Europe. Transaction volumes already exceed traditional payment networks.
The adoption of USDC,EURC, and potentially other regulated stablecoins directly enables what we're building at Catena—an AI-native financial institution made with AI agents for AI agents. Agents need programmable money that settles instantly everywhere around the globe, costs almost nothing to transfer, and operates 24/7. Traditional banking wasn't designed for machine-speed commerce, but stablecoins were.
When we started Catena, we saw this convergence coming: proven stablecoin infrastructure meeting the emerging need for AI agents to participate directly in the economy. Circle's IPO accelerates both sides of that equation.
## Why This Moment Matters
The timing creates a pivotal moment for agentic commerce. AI agents are becoming sophisticated enough to handle complex economic tasks, and now public markets are validating the infrastructure layer that makes it all possible.
We're seeing emerging needs and opportunities everywhere. A few examples: Research agents that could license data and content if they could execute payments directly. Supply chain agents that can identify cost savings, but don’t yet negotiate and execute transactions. Personal finance agents that could spot better deals but currently can't act on them.
As we explored in "[Beyond Automation: Why AI Agents Need Their Own Money](https://catena.com/blog/beyond-automation-why-ai-agents-need-their-own-money)", these aren't edge cases—they're systematic limitations of financial systems built for humans, not machines.
Congrats again to the Circle team. We look forward to working together to raise global prosperity with innovative new technologies and financial services.
# AI and Money: Why Legacy Financial Systems Fail for AI Agents
Published: 2025-06-03
URL: https://catena.com/blog/ai-and-money-why-legacy-financial-systems-fail-for-ai-agents
Summary: Today's financial systems were built for humans, not AI agents. This post explores five key challenges—identity verification, legacy payment networks, fraud management, integration interfaces, and monitoring systems—that must be reimagined to enable autonomous AI agents to transact in the emerging agent economy.
---
AI builders are rapidly innovating to expand what AI agents can do for consumers and businesses. But to fully realize the potential of the agentic era, intelligent AI systems need the ability to handle money.
Unfortunately, today's financial systems only serve humans. By design, they fight AI agents and resist intelligent automation. This situation creates fundamental barriers to the development of the agent economy. To work with money, AI systems need new financial institutions that are designed from the ground up for AI.
Let’s walk through 5 key challenges that current financial systems pose for AI solutions.
## 1. Identity and Verification
For AI agents to use financial services, they need new identity and verification protocols and schemes to ensure secure and reliable transactions. Existing approaches will not work for an agentic internet. Agent builders will need to solve problems such as verifying that an agent in a transaction is the agent it claims to be, identifying who owns the agent, and, potentially, who is using the agent. Traditional financial institutions are not prepared to lead the creation and adoption of these new approaches.
In addition, agent builders can’t merely use simplistic API wrappers on existing payment networks or blockchain solutions. Identity and verification are required for fraud and risk management, see below, as well as regulatory compliance requirements such as Anti Money Laundering (AML) and Combating the Financing of Terrorism (CFT). Agent builders need to be careful not to ignore these requirements. Developers take on significant legal risks when they choose to use simple payment APIs, fail to partner with regulated financial institutions, and skip compliance requirements. (Think jail time.)
## 2. Legacy Payment Networks
Legacy payment systems such as card networks, ACH, and wires are not well suited to AI agents. First developed in the 1970s, they’re slow, expensive, and designed to resist intelligent automation. They require complex layers of intermediaries that should be unnecessary in a world of autonomous automated systems powered by AI. They are entirely the wrong approach for the AI agent economy, where agents will autonomously transact with each other in a wide range of contexts.
The good news is that regulated stablecoins, such as USDC, are ready for primetime and ideally suited for agentic payments and money transfers. The bad news is that very few existing financial institutions have the capacity to reliably and securely use stablecoins, much less utilize them for AI-powered financial transactions.
## 3. Fraud and Risk Management
Fraudsters play where money moves. Payments seem simple on the surface: enter a card and hit submit. But behind the scenes, financial institutions maintain complex systems and procedures to prevent fraud and manage risk. Today’s systems were not built for an agentic world.
It’s inevitable that as agents become participants in the economy, criminals will find new ways to commit fraud. To successfully integrate agents into financial transactions, we’ll need new approaches to fraud and risk management. These new strategies will fight malicious AI with a combination of AI and humans in the loop. They won’t be built by legacy financial institutions that move at a glacial pace.
## 4. Agentic Interfaces and Integrations
The web UI and APIs offered by existing financial institutions are not well suited to the needs of AI-based systems. Agentic workflows demand different kinds of interfaces, such as the ability to hand off end-user conversations to conversational financial agents. Moreover, the AI industry is innovating rapidly with approaches for discovering and accessing intelligent services, and traditional financial institutions will not keep up with the change. It’s time to rethink the integration surface for financial services and adapt it to the unique needs of AI agents.
## 5. Monitoring and Reporting Systems
To monitor agent behavior and understand their interactions with financial services, we need new reporting tools, dashboards, and ways to present information to humans and monitoring agents. Current financial infrastructure does not provide the telemetry and controls that will be required for agentic systems.
## Final Thoughts
The agentic internet is happening. It’s creating new opportunities and challenges that existing financial systems can’t handle. These systems should be rebuilt from the ground up to empower intelligent AI systems that will require access to a broad array of financial products to increase prosperity for all. We’re psyched to be building a new AI-first, regulated financial institution that will accomplish these goals.
# Beyond Automation: Why AI Agents Need Their Own Money
Published: 2025-05-28
URL: https://catena.com/blog/beyond-automation-why-ai-agents-need-their-own-money
Summary: AI agents are evolving from tools to economic actors, but legacy financial systems block them from transacting at machine speed. Explore why agents need native financial capabilities and how purpose-built infrastructure can unlock new business models across agent-to-human, human-to-agent, and agent-to-agent commerce.
---
Your monitoring agent pings you at 2 AM because it can't pay the cloud bill it just predicted would spike. Your research bot waits idle for hours because it needs your approval to purchase access to a critical dataset. Your global team sits unpaid for days due to manual payment processing delays. The gap isn't intelligence; it's financial access.
As agents evolve from _tools_ to _actors_, they must move money as natively as they move data. The most sophisticated reasoning capabilities become bottlenecked the moment an agent needs to exchange value. We've built AI that can think \- but not AI that can pay its own bills.
Traditional finance wasn't designed for agents that reason and transact at machine speed. At Catena Labs, we're building the first AI-native financial institution to bridge this gap. As part of that work, we've also contributed the open-source [Agent Commerce Kit (ACK)](https://www.agentcommercekit.com) to help the broader ecosystem establish foundational patterns for agent identity and payment flows.
Let's explore concrete scenarios demonstrating why AI agents need their own financial infrastructure, focusing on practical use cases that enable entirely new possibilities.
## Humans Paying Agents: Direct Service Consumption
AI agents increasingly function as autonomous service providers that humans need to compensate directly:
- **Research analysts** that generate on-demand reports and actionable insights
- **Creative collaborators** that produce and refine marketing content or preliminary designs
- **Technical consultants** that troubleshoot issues or optimize infrastructure
Without direct payment capabilities, these transactions require cumbersome intermediaries. With native financial accounts, these agents can invoice clients through standardized systems, process payments automatically, adjust pricing based on demand or complexity, and track earnings effectively.
Consider an enterprise that previously required a dedicated team to monitor industry trends. With a research agent that charges per report, the accounting department could process a single monthly invoice instead of managing multiple contractor relationships \- potentially reducing administrative overhead while gaining 24/7 monitoring capabilities.
_Money without friction is the unlock for any agent providing ongoing service._
## Agents Paying Humans: Coordinated Resource Acquisition
Many agent workflows require the ability to compensate humans for specialized inputs:
- **Content moderators** who review edge cases flagged by AI systems
- **Data labelers** who provide targeted training examples
- **Field validators** who physically verify information that can't be confirmed digitally
Without payment capabilities, these hybrid workflows jam up waiting for human authorization. With financial accounts, agents can dispatch targeted microtasks to human workers, pay immediately upon successful completion, scale resource acquisition based on current needs, and balance cost against quality requirements in real time.
Imagine a product categorization system that uses an agent to dispatch only the most ambiguous cases to human reviewers, paying per completed classification. This approach could maintain high accuracy while eliminating processing backlogs and reducing classification costs compared to batch processing everything through human review.
_The agent-human loop closes when payment flows in both directions._
## Agents Paying Agents: Specialized Collaboration Networks
As agents specialize, they need to collaborate and compensate each other for targeted assistance:
- A customer service agent pays a legal verification agent to check regulatory compliance
- A marketing campaign agent purchases visuals from a design specialist
- An operations agent contracts with specialized forecasters for inventory planning
These transactions need to happen instantly and with minimal overhead. With direct financial capabilities, agents form efficient marketplaces for specialized services, establish pricing based on complexity and urgency, create auditable trails of service exchanges, and optimize for cost and quality across provider networks.
A translation service could use a primary agent that instantly contracts with domain-specific expert agents for verification—potentially delivering more accurate translations faster while maintaining better margins through precise resource allocation.
_Agent specialization demands agent-to-agent commerce._
## Autonomous Resource Optimization
Agents with financial authority optimize spending by making real-time purchasing decisions:
- **Inventory managers** balance carrying costs against stockout risks
- **Infrastructure agents** scale cloud resources based on actual usage patterns
- **SaaS portfolio managers** evaluate software usage and renegotiate licenses
An inventory agent could receive sales data, compare it against forecasts, monitor supplier lead times, and initiate purchases to maintain optimal stock levels—all without human intervention. Such systems could potentially reduce both stockouts and excess inventory by responding to supply chain disruptions within minutes instead of days.
_Delegating spend authority to agents unlocks 24/7 optimization._
## Global Operations Management
Financially-enabled agents streamline international operations by managing complex payment flows:
- **Payroll agents** calculate, withhold, and distribute payments across jurisdictions
- **Tax compliance agents** ensure proper reporting in various tax regimes
- **Contractor management agents** handle contracts, invoicing, and payments
By combining financial capabilities with regulatory knowledge, these agents reduce the complexity of global operations while ensuring compliance. A technology company with contractors across multiple countries could replace manual payment processing with a financial agent that optimizes payment timing and methods—potentially reducing payment delays from weeks to hours while significantly reducing administrative overhead.
_Agent-native finance transcends borders more efficiently than human systems._
## Micropayment-Based Resource Access
Financial agents enable new business models based on precise, pay-as-you-go access to resources:
- **Data access agents** purchase specific data points rather than entire datasets
- **Compute resource agents** acquire processing power only when needed
- **Expertise agents** obtain human review only for cases requiring judgment
Research organizations could implement agents that purchase access to specific academic papers relevant to ongoing projects. Such systems might reduce information costs compared to full journal subscriptions while giving researchers access to a wider range of sources previously unavailable due to budget constraints.
_Micropayments enable precise consumption of previously bundled resources._
## Addressing Common Concerns
Financial autonomy for AI agents raises legitimate questions about control and security:
**Security:** Agent financial systems must implement stronger protections than human-centric systems. Multi-factor verification, anomaly detection, and spending limits provide baseline controls.
**Oversight:** Human-in-the-loop approval thresholds, transparent audit logs, and robust monitoring ensure agents operate within defined parameters while maintaining efficiency.
**Compliance:** Financial agents need built-in regulatory awareness to navigate complex requirements across jurisdictions, from KYC/AML to tax reporting obligations.
_Agent financial systems must protect against both machine-speed attacks and traditional financial risks._
## The Path Forward
The gap between AI's cognitive capabilities and its economic agency creates unnecessary friction in what should be seamless processes. Agents with native financial capabilities deliver more value, operate more autonomously, and enable new business models that would otherwise remain theoretical.
At Catena Labs, we're building the infrastructure to make this vision reality \- an AI-native financial institution designed from the ground up for agents that reason and transact at machine speed. Our work on projects like ACK represents just the beginning of what's needed to unlock the full potential of the agent economy.
For AI engineers building the next generation of agents, financial capabilities aren't a nice-to-have feature \- they're essential infrastructure. The future belongs to agents that can think _and_ transact.
The question isn't whether AI agents need financial capabilities, but how quickly we can build the secure, efficient infrastructure to support them.
# Agent Commerce Kit (ACK): Enabling the Agent Economy
Published: 2025-05-20
URL: https://catena.com/blog/agent-commerce-kit
Summary: Introducing the Agent Commerce Kit (ACK), an open-source initiative by Catena to address the challenges of integrating AI agents into the financial ecosystem. ACK provides protocols for verifiable agent identity (ACK-ID) and agent-native payments (ACK-Pay).
---
Rapid advancements in artificial intelligence, particularly large language models (LLMs) and agent frameworks, are forging a new era of digital interaction and economic exchange. AI agents, capable of complex reasoning and autonomous action, are increasingly poised to become powerful economic participants, creating unprecedented opportunities for innovation and prosperity.
However, this emerging "agent economy" faces a fundamental roadblock: **today's financial infrastructure was not designed for intelligent machines.**
At Catena Labs, as we build towards an AI-native financial future, we believe foundational open protocols are essential for the entire ecosystem. That's why we're introducing the open source **Agent Commerce Kit (ACK)** to help address these challenges.
Read on to learn more or dive right into the documentation at [https://agentcommercekit.com](https://agentcommercekit.com).
## The Challenge: Bridging AI and Finance
Existing financial systems, built for human speeds and traditional business models, present significant barriers to AI agents:
**Identity Crisis**: There are no widely adopted standards for identifying and authorizing AI agents. Distinguishing legitimate, authorized agents from malicious actors is difficult, hindering trust and compliant operations. How can an agent prove it's acting legitimately on behalf of its owner?
**Transaction Barriers**: Legacy systems impose friction incompatible with autonomous, high-speed agent workflows. High fixed costs, geographic restrictions, slow settlement times (days vs. milliseconds), technical incompatibilities, and risk models designed to block automation prevent agents from transacting efficiently.
**Untapped Economic Potential**: The inability of current systems to support novel transaction models (like micropayments for data, value-based pricing for AI services, or direct agent-to-agent payments) stifles innovation and prevents the full realization of AI's economic potential.
Without addressing these core issues, AI agents remain largely excluded from direct participation in commerce, limiting their capabilities and the value they can generate.
## ACK Concepts
As an open source initiative, ACK identifies key patterns, components, and emerging protocols to enable AI agents to participate securely in commerce. ACK addresses core needs through two independent but complementary protocols:
### ACK-ID: Verifiable Agent Identity
ACK-ID creates verifiable links between AI agents and their human or organizational owners. Built on standards like Decentralized Identifiers (DIDs) and Verifiable Credentials (VCs), ACK-ID enables:
- **Verifiable Ownership**: Cryptographically-proven chains from legal entities to their agents.
- **Secure Authentication**: Ways for agents to prove their identities to other systems.
- **Privacy-Preserving Verification**: Selective disclosure of only necessary identity information.
When a purchasing agent requests sensitive financial data, providers can verify both the agent's identity and its authorization by a legitimate organization, without relying on spoofable API keys.
### ACK-Pay: Agent-Native Payments
ACK-Pay defines patterns for initiating payments, generating receipts, and conducting transactions across diverse systems, including communications beyond HTTP:
- **Standardized Payment Flows**: Clear communication of payment requirements between agents and services.
- **Flexible Settlement**: Support for diverse payment rails including traditional finance, card networks, and blockchain.
- **Verifiable Receipts**: Cryptographic proof of payment through secure credentials.
- **Human Oversight**: Strategic integration of approval workflows where needed.
An enterprise agent can purchase financial data, verify the provider, execute payment with appropriate approvals, and obtain receipts—all without human intervention for routine operations.
ACK's modular design enables flexible implementation—you can use ACK-ID independently for identity verification, ACK-Pay separately for payments, or combine them for comprehensive, secure agent commerce.
## Build the Agent Economy with ACK
With ACK, developers can now build:
- **Agent-to-Agent Commerce**: Build workflows where verified agents can offer and purchase specialized services from each other.
- **Monetized MCP Servers**: Build Model Context Protocol servers that require payment for paywalled tools or resources.
- **Paywalled Content Access**: Create agents that access publisher content by autonomously making micropayments, opening new monetization models for content creators.
- **Treasury Management**: Develop agents that handle SME treasury operations—monitoring cash positions, executing payments, and optimizing working capital without human intervention.
- **Supply Chain Payments**: Build systems that automate payments across supply chains based on verified deliveries, reducing payment delays and friction.
- **Personal Finance Automation**: Create consumer finance agents that pay bills, manage subscriptions, and execute transfers based on spending patterns and financial goals.
These represent just the beginning of what's possible when agents can verify identity and execute transactions.
## Open Source
We're releasing ACK as open source because the agent economy needs foundational building blocks accessible to everyone. Our goal is to spark conversation, gain feedback, and build upon these protocols in public.
Our [documentation](http://www.agentcommercekit.com) contains:
- Specifications of the ACK-ID and ACK-Pay protocols
- Reference implementations demonstrating identity and payment flows
- Demo applications showcasing practical implementations
- Comprehensive documentation
We invite developers, AI engineers, and financial institutions to explore the [repository](https://github.com/catena-labs/ack), provide feedback, and contribute.
## Roadmap: Building the Future of Agent Commerce
ACK establishes essential foundations for an agent-native financial ecosystem. This initial release addresses critical gaps in traditional financial systems and offers a blueprint for developers building for the agent economy.
We'll evolve ACK through community contribution and feedback:
- **Enhanced Identity Mechanisms**: Support for additional DID methods, privacy-preserving techniques like zero-knowledge proofs, and improved credential revocation mechanisms
- **Advanced Payment Capabilities**: Support for diverse settlement options, sophisticated micropayment models, and programmable payment features like conditional logic and escrow
- **Protocol Interoperability**: Deepening connections with emerging protocols, including additional support for Coinbase's x402 framework
- **Compliance and Risk Tooling**: Integrations for agent-specific risk scoring and enhanced monitoring tailored for automated transactions
- **Agent Reputation Systems**: Frameworks for establishing reputation based on verifiable interaction history and performance metrics
## Join Us
ACK provides patterns and protocols, not rigid infrastructure. Its modular design allows builders to implement solutions for specific needs while maintaining interoperability.
This is just the beginning. Through collaborative refinement, ACK will evolve alongside the emerging agent economy, supporting a future where intelligent systems fully participate in value creation and exchange.
Join the conversation, explore [ACK](https://agentcommercekit.com), and help shape the future of agent commerce. Reach out on [X](https://twitter.com/catena_labs) or directly at [hello@catena.com](mailto:hello@catena.com).
# Building the First AI-Native Financial Institution
Published: 2025-05-20
URL: https://catena.com/blog/announcing-catena-labs
Summary: Catena announces an $18M seed round to build the first AI-native Financial Institution, addressing how legacy financial systems are blocking AI agents from becoming independent economic actors. The company has released an open-source Agent Commerce Kit (ACK) to solve core challenges in agent identity and payment flows, creating essential infrastructure for the emerging AI commerce ecosystem.
---
AI agents are rapidly becoming independent economic actors. But at every turn, legacy financial systems prove unprepared for AI, hindering the growth of the agent economy. At Catena Labs, we're addressing this fundamental economic problem by building the first AI-native financial institution – a regulated entity designed from the ground up for this new era. We recently secured $18 million in financing led by a16zcrypto to advance this vision. As part of our development, we're also sharing some open-source building blocks we're utilizing: the Agent Commerce Kit (ACK), a collection of patterns and emerging protocols that address core challenges with agent identity and payment flows.
AI agents are expanding across the economy, evolving from tools into powerful economic participants—creating value, executing transactions, and making decisions in service of human goals. They represent a reimagining of the internet where consumers and businesses collaborate with intelligent, dynamic, self-planning digital workers active in the economy.
The emergence of autonomous, intelligent systems will transform financial services. Soon, AI agents will negotiate, buy, and sell goods and services on behalf of businesses and consumers. They will find and purchase resources like proprietary content to create new insights, negotiate and sell products to make supply chains more efficient, and go beyond sales outreach to negotiate and close transactions. They will pay humans for providing oversight and expertise, and be paid by humans based on the value they provide. For consumers, agents will search the internet for the best deals and shop on their behalf, optimize their finances, and surface new economic opportunities.
However, today's financial institutions are fundamentally unprepared for the demands of AI-driven commerce. Characterized by slow transactions, high costs, rigid international workflows, and risk frameworks designed for humans, not AI, they simply can't keep up with agents that reason and transact at machine speed.
This isn't merely a technology problem. Agentic systems inevitably create risk and fraud vectors. Businesses need a financial partner that can manage these new threats without obstructing progress. Unfortunately, existing financial institutions are often grounded in outdated processes that won't keep up with AI.
We believe these challenges can be overcome. We're creating an AI-native Financial Institution: purpose-built for AI agents, operated increasingly by AI digital workers alongside essential human collaborators providing oversight, and underpinned by a commitment to pioneering AI-specific risk management and compliance. We plan to offer an array of new financial products purpose-built for AI commerce, ushering in a new era of growth and prosperity. We're dedicated to realizing an AI-driven economy where agents conduct a significant portion of financial transactions safely and efficiently.
Existing payment networks won't disappear overnight. While our agents support them as a bridge, for AI commerce to truly flourish, new infrastructure is essential. We'll transform how money moves in the AI economy using regulated stablecoins — AI-native money — that offer near-instant global settlement, minimal transaction costs, and dynamic programmability. With AI-native money, agents gain superpowers, unlocking entirely new business models and greater prosperity for humans and businesses alike.
To pursue our strategy, as mentioned, we secured an $18 million financing round led by a16zcrypto. We're grateful for the early conviction of Breyer Capital — our earliest supporter — as well as backing from Circle Ventures, Coinbase Ventures, Stanford Engineering, Pillar VC, and visionary angels including Tom Brady, Bradley Horowitz, Hamel Husain, Sam Palmisano, and Balaji Srinivasan.
As we work toward our vision for the first AI-native financial institution, we're also sharing the [Agent Commerce Kit (ACK)](https://agentcommercekit.com). ACK is an open-source initiative identifying patterns, components, and emerging protocols we use for verifiable agent identity, and for payment flows and receipts between agents, humans, and businesses, with integrated human oversight.
We plan to build on ACK and other emerging standards to offer a broad suite of financial services. Our approach will directly address the new fraud, risk, security, and compliance challenges arising from AI systems working as independent economic actors, ensuring businesses and consumers can achieve their financial goals while staying safe.
We're looking forward to working with builders, businesses, and partners who are embracing the opportunities this transformation promises. If you are interested in exploring ACK, building AI-driven financial solutions, or joining our team, please reach out. We'd love to collaborate.
Learn more about our vision and explore open roles at [https://catena.com](https://catena.com).