Two strategies, one problem: why your AI and stablecoin roadmaps should merge

Hero Image
  • Blog
  • November 09, 2026

There is a familiar pattern playing out in strategy and technology functions across industries right now: One team is building out AI capabilities such as automating payment initiation, optimizing working capital flows, enabling autonomous vendor settlement. Another team, usually in tech or a side practice, is discerning how to use, manage, and optimize digital assets. Both teams are making real progress in their proof of concepts and traveling further down their respective journeys.

And in most organizations, they are not talking to each other.

Alternative Text

This blog was produced in collaboration with Hedera.

Ilana Golbin Blumenfeld

Ilana Golbin Blumenfeld

Principal, Trust AI, PwC US

Kelly Guinan

Kelly Guinan

Director, PwC US

Aditi Lyall

Aditi Lyall

Principal, Digital Product Management and Blockchain, PwC US

Caroline McIlroy

Caroline McIlroy

Director, Digital Assets, PwC US

Key takeaways:

  • AI and stablecoin strategies should be developed together. Agentic payments depend on the data, access rights and controls created by the payment infrastructure underneath them.
  • Stablecoin adoption can change the data environment AI relies on. As settlement shifts on-chain, signals, enrichment layers and access rights may change in ways that affect model performance and governance.
  • Separate roadmaps can create technical and operating-model gaps. Independent vendor, data and control decisions can lead to fragmented infrastructure and costly integration later.
  • An integrated payments architecture can create strategic advantage. Coordinating AI and stablecoin decisions can strengthen interoperability, data rights, controls and signal quality across traditional and on-chain rails.
  • Strategy leaders should act before standards harden. Decisions on vendors, data access, governance and interoperability made now may shape how effectively organizations scale agentic payments in the future.

The problem

The framing of "AI strategy" and "digital asset” or “stablecoin strategy" as parallel workstreams implies they operate on independent tracks that can be integrated later. In reality, each one changes the foundation the other depends on. The AI roadmap relies on data, access rights, and controls built around traditional payment infrastructure. The stablecoin roadmap is changing that infrastructure, along with the data it produces, who can see it, and the rules that govern it.

If your AI systems are trained and validated on traditional payment flows, but your settlement infrastructure shifts meaningfully toward stablecoins, you face a significant infrastructure and governance problem. Inputs change, performance degrades, and because the shift happens gradually, the degradation may not be obvious until the consequences are material.

Because separate teams own each workstream, no one is positioned to see when a decision in one constrains the other. Organizations building on assumed technology rather than dynamic or composable stacks will carry significant technical debt when the next wave of innovation arrives.

What each technology enables in isolation

Agentic AI in payments shifts the unit of action from a human-initiated transaction to a continuous, machine-orchestrated flow. Rather than a treasury analyst approving an FX hedge or a procurement system triggering a purchase order, agentic systems monitor conditions, assess risk parameters, negotiate terms, and execute. This all happens without waiting for human input at each step. The value is speed, scale, and consistency, with risk managed through guardrails that governance teams define and monitor.

When deployed on traditional rails such as card networks, ACH, SWIFT, agentic AI works within a familiar data environment. The signals are structured, the intermediaries are known, and the compliance frameworks are established. Organizations can build meaningful automation here today.

Alternative text

Stablecoin-based settlement offers a different set of capabilities. Programmable logic is embedded in the transaction itself, meaning that settlement can happen in seconds rather than days once payment conditions are fulfilled. Atomic delivery-versus-payment can eliminate the lag between obligation and finality, accelerating cash and reducing treasury operations workload. For cross-border flows, high-volume micropayments, and supply chain finance, the efficiency gains are material.

The data environment, however, looks very different from a card swipe or a wire transfer. Stablecoin payments leave a permanent, verifiable record of every transfer, but they carry less of the context that traditional payments include, such as who is on the other side, what the payment is for, and how to reverse it. There are tools in traditional payment systems today that normalize, enrich, and govern access to payment data, and equivalent tools exist in the blockchain ecosystem, but few span both. Verifiable credentials (VCs) are one potential path to shared trust for agentic signals by presenting a means for cryptographically securing agentic payments. The standards for what get recorded, who can read it, and under what conditions are still being written.

Alternative text

What happens when you build them in isolation

This is where strategic divergence starts to matter.

An agentic payment system that runs on traditional rails makes decisions based on a specific data architecture. It depends on the signals it can access, how quickly those signals arrive, and the enrichment layers that sit between the raw transaction and the model. That architecture has been built over decades, and it reflects long-standing assumptions about what payment data looks like.

Stablecoin settlement challenges those assumptions at a very large scale. The available signals change, some enrichment layers may disappear, and the access rights your agentic system depends on, including data-sharing agreements, network rules, and API contracts, were negotiated for a different infrastructure.

This may also happen sooner than broader market adoption signals suggest. Instant settlement, digital-native assets, and programmable logic make stablecoins a natural fit for agentic payment use cases, so the two roadmaps may meet well before stablecoins reach scale across the payments market.

Alternative text

These are not just technical decisions for engineers. They are being shaped through standards bodies, regulatory consultations, network governance processes, and vendor contracts. Organizations that treat stablecoin infrastructure as a treasury issue and AI as a technology issue risk being underrepresented in those conversations because they are not showing up with an integrated point of view.

When you build the roadmaps in isolation, the failure modes are specific.

Vendor fragmentation. Your AI infrastructure vendor may have built data pipelines around traditional payment signals. Your stablecoin custody and settlement provider may have built around on-chain data. If these decisions are made independently, you can end up with two data environments that do not communicate well with each other. That creates significant integration costs when you try to run agentic logic across traditional and stablecoin rails at the same time.

Operational model mismatch. Agentic AI requires clear escalation logic, defined risk tolerances, and audit trails that satisfy compliance requirements. Stablecoin settlement operates in a regulatory environment that is still evolving. If your operating model for AI-driven payments is built only around traditional payment compliance frameworks, it will not translate cleanly to on-chain settlement. The teams that own risk and controls need to design for both environments together, not one after the other.

Access rights that expire. Many of the data-sharing arrangements that give agentic systems their signal quality are embedded in network rules or bilateral agreements tied to specific settlement infrastructure. As stablecoin volume grows, new access frameworks will emerge. Organizations that help define those frameworks will have structural advantages. Those that arrive later will have to accept terms set by others.

The window to shape those standards is open. Institutional stablecoin volume is growing, but it has not yet reached the scale where the architecture becomes fixed. When architecture is more fixed, organizations with integrated roadmaps will have stronger data rights, better interoperability access, and higher-quality model signals that others will struggle to replicate. Organizations that do not act now may find themselves negotiating for access to intelligence they helped create but do not control.

What happens when you do it correctly

When the roadmaps are built together, the same decisions that create failure modes in isolation become sources of advantage.

Alternative text

Shared data architecture. Your AI infrastructure vendor and your stablecoin custody and settlement provider are selected together, so their data pipelines work across traditional payment signals and onchain data. That avoids the significant integration costs of running agentic logic across traditional and stablecoin rails at the same time, and agentic systems keep the signal quality they depend on as settlement shifts.

One operating model. The teams that own risk and controls design for both environments together, not one after the other. Escalation logic, defined risk tolerances, and audit trails that satisfy compliance requirements carry over cleanly to onchain settlement.

Access rights that grow. Organizations that show up with an integrated point of view help define the new access frameworks. They gain structural advantages: stronger data rights, better interoperability access, and higher-quality model signals that others will struggle to replicate.

The result is a single roadmap in which each use case can run on the best rail for the job, with the data, controls, and access rights already in place underneath it.

The questions strategy leaders should be asking now

The argument is not that you need to solve this integration before making progress on either roadmap. It is that the decisions made over the next twelve to eighteen months will either preserve or limit options that will matter at scale. That means leaders should ask a focused set of questions while the architecture is still flexible.

Data architecture: What signals does your agentic payment strategy depend on, and which of those signals exist in the stablecoin data environment? Where are the gaps? Who is working to fill them through standards, data products, or enrichment layers? Are you part of those conversations?

Vendor strategy: When you evaluate AI infrastructure vendors and stablecoin settlement providers, are you assessing how well they work together, not just how strong they are individually? What does the data handoff look like when a payment initiated by an AI agent settles on-chain?

Operating model and controls: How does your risk and compliance framework need to evolve to govern agentic decisions across both traditional and stablecoin rails? Who owns that evolution? Are they working from one integrated brief or two separate ones?

Standards and influence: Where are the key governance conversations happening around stablecoin data standards and interoperability? Are your technology, treasury, risk, and strategy teams represented with a coherent view of what the intelligence layer should become?

Timing: When do you expect stablecoin settlement to reach institutional scale in your industry? Is your integration roadmap designed to get ahead of that inflection point, or will it lag behind it?

The strategic reframe

The organizations with structural advantages in the agentic payment era will not necessarily be the ones that move fastest on AI or fastest on stablecoins. They will be the ones that understand these are part of the same infrastructure bet and make their architecture, vendor, and governance decisions accordingly.

The payment rail is changing. The data layer that will run alongside it is being designed now. The intelligence that agentic systems will rely on for the next decade is being embedded in decisions that may look like technical standards work, treasury vendor selection, or infrastructure procurement.

They are not just technical or procurement choices. They are strategy decisions.

Leaders who treat them that way, and who bring their AI and stablecoin roadmaps into the same room before the architecture hardens, will build advantages that are difficult to replicate.

Contact us

Ilana Golbin Blumenfeld

Ilana Golbin Blumenfeld

Principal, Trust AI, PwC US

Kelly Guinan

Kelly Guinan

Director, PwC US

Aditi Lyall

Aditi Lyall

Principal, Digital Product Management and Blockchain, PwC US

Caroline McIlroy

Caroline McIlroy

Director, Digital Assets, PwC US

Follow us

Required fields are marked with an asterisk(*)

Your personal information will be handled in accordance with our Privacy Statement. You can update your communication preferences at any time by clicking the unsubscribe link in a PwC email or by submitting a request as outlined in our Privacy Statement.

Hide