Weekly reports

The Week in One Paragraph

Week 39 of 2026 was defined by two converging forces: the industrialisation of agentic payments infrastructure, and Stripe's aggressive move to position itself as the economic layer for AI commerce. Ant International, Visa, and Mastercard published a joint Know-Your-Agent framework, while the same networks simultaneously rolled out live agentic payment products — signalling that standards and infrastructure are now developing in parallel rather than in sequence. Stripe's $7.5 billion acquisition of OpenRouter extends its ambitions well beyond payment processing into AI model routing, a bet that the company best placed to meter AI compute spend will also capture the payments layer beneath it. Against this backdrop, the European Parliament's approval of the digital euro Regulation and the opening of a merchant pilot provide the clearest signal yet that a state-backed digital currency will be a structural feature of European payments by the end of the decade. Visa's simultaneous expansion across stablecoins, onchain lending, passkeys, and dispute management reflects a network executing on multiple fronts at once, though the volume of announcements makes it difficult to assess which will achieve meaningful adoption.

Top Developments

1. Ant International, Visa, and Mastercard Publish Joint Know-Your-Agent Framework

What happened: On 10 September 2026, Ant International, Visa, and Mastercard announced a collaboration to develop an interoperable Know-Your-Agent (KYA) framework for AI agent-initiated payments. The framework draws on each network's existing protocols — Visa's Trusted Agent Protocol, Mastercard's Verifiable Intent, and Ant's Agentic Mobile Protocol — and references a regulatory SAFR framework. Concurrently, ReconArt integrated with Solana for stablecoin and agentic payment reconciliation, and Paystand launched an agentic finance suite for B2B payments.

Why it matters: Identity and accountability for non-human payment initiators is the foundational unsolved problem in agentic commerce. A cross-network KYA standard, if adopted broadly, would define which agents can transact, on what terms, and who bears liability when they err — questions that will determine the commercial architecture of AI-driven payments for years. PSPs and merchants need to track whether this framework becomes a genuine interoperability standard or fragments into competing proprietary implementations.

Companies: Ant International, Visa, Mastercard, ReconArt, Paystand.

Impact: 68 | Confidence: 72

2. Card Networks Deploy Live Agentic Payment Products Across Multiple Markets

What happened: Mastercard's Agent Pay has rolled out globally with issuing bank participation, including a partnership with Alchemy to integrate AgentCard with one-time-use tokenised credentials. Visa completed a live proof-of-concept with Nuvei, Arvato Systems, and fashion brand Kings and Priests. Stripe is powering Meta's Muse AI shopping agent. Paytm and BharatPe are deploying AI agents for enterprise and merchant support in India. Mastercard's Verifiable Intent specification was co-developed with Google and open-sourced through the FIDO Alliance.

Why it matters: The shift from framework announcements to live deployments is significant. Issuing banks now face near-term decisions about whether to participate in Agent Pay and equivalent schemes, and on what liability terms. Merchants integrating AI shopping agents — or whose customers use them — need to understand how delegated payment authority interacts with their existing fraud, returns, and dispute workflows. The open-sourcing of Verifiable Intent through FIDO is a deliberate move to accelerate adoption and reduce the risk of proprietary lock-in becoming a barrier.

Companies: Mastercard, Visa, Nuvei, Alchemy, American Express, Fireblocks, PayPal, Stripe, Checkout.com, Paytm, BharatPe.

Impact: 68 | Confidence: 65

3. Stripe Acquires OpenRouter for $7.5 Billion, Entering AI Model Infrastructure

What happened: Stripe finalised the acquisition of OpenRouter, an AI gateway and model marketplace providing access to 400+ AI models from 80+ providers, for approximately $7.5 billion, with $1.5 billion allocated to founders. OpenRouter's core capability is optimising token usage and routing across providers, having analysed 55 trillion tokens of spend. The deal follows Stripe and Advent International abandoning a $53.4 billion pursuit of PayPal in late August 2026.

Why it matters: This is not a payments-adjacent acquisition — it is a direct entry into AI infrastructure. Stripe is positioning itself to capture spend at the model layer, not just the payment layer, as AI companies become a major and growing customer segment. For competing PSPs, the question is whether Stripe's combined payment-plus-AI-routing proposition creates a switching cost that is structurally difficult to replicate. The abandoned PayPal bid, read alongside this deal, suggests Stripe concluded that building AI infrastructure was a higher-return use of capital than acquiring consumer payments scale.

Companies: Stripe, OpenRouter, Advent International, PayPal.

Impact: 68 | Confidence: 80 (averaged across two corroborating entries with scores of 82 and 78)

4. European Parliament Approves Digital Euro Regulation; Merchant Pilot Opens

What happened: The European Parliament approved the Regulation for the digital euro, with issuance referenced for 2029. The Parliament and Council have both adopted negotiating positions on the Single Currency Package, which includes a proposed Regulation on the legal tender status of euro banknotes and coins. The Eurosystem has separately opened a call for e-commerce and m-commerce merchants to participate in a digital euro pilot.

Why it matters: Parliamentary approval moves the digital euro from policy debate to legislative process. For PSPs and acquirers operating in the euro area, the 2029 horizon is now a planning assumption rather than a speculation — acceptance infrastructure, wallet integrations, and settlement arrangements will need to be scoped. The merchant pilot call is the first concrete operational step: firms that participate will have an informational and integration advantage. The legal tender package running in parallel matters for cash-handling obligations and the broader monetary framework within which digital euro sits.

Companies: European Central Bank, Eurosystem.

Impact: 65 | Confidence: 65 | Note: the 2029 issuance date is flagged in the source material as possibly conditional rather than confirmed policy.

5. Visa Expands Across Stablecoins, Onchain Lending, Passkeys, and Dispute Management

What happened: Visa announced integration of onchain lending with VisaNet settlement data to support stablecoin-linked card programs and fintech working capital access, as part of its broader Visa Stablecoin Platform. Separately, Visa launched Payment Passkey with In-Solutions Global for passwordless authentication, expanded cardless payment solutions with PhonePe across 14 markets including Cross Border Scan to Pay, and extended real-time dispute management tools with payabl. for UK and EU merchants. Visa also faces merchant pushback on a proposed interchange settlement with Mastercard, with critics arguing fee reductions are insufficient.

Why it matters: The breadth of Visa's simultaneous announcements reflects a network competing on multiple dimensions at once — digital assets, authentication, cross-border, and post-purchase services. The onchain lending integration is strategically notable: connecting VisaNet settlement data to stablecoin-linked working capital creates a data moat that is difficult for non-network competitors to replicate. The merchant fee settlement friction is a persistent risk to network relationships and warrants monitoring.

Companies: Visa, Checkout.com, payabl., PhonePe, In-Solutions Global, Mastercard.

Impact: 68 | Confidence: 55 (the merchant settlement item carries a confidence score of 45 — treat with caution)

Themes This Week

Agentic payments is moving from concept to infrastructure simultaneously at the standards and product layers. Three of the five top stories involve AI agents initiating payments. What is unusual this week is that standards work (KYA framework) and live deployments (Agent Pay, Visa-Nuvei PoC, Stripe-Meta Muse) are happening at the same time. Historically, infrastructure standards precede products. The inversion creates risk: proprietary implementations may entrench before interoperability norms are settled, and liability frameworks for agent-initiated errors remain unresolved.

Stripe is redefining what a payments company is. The OpenRouter acquisition, combined with powering Meta's AI shopping agent, signals that Stripe is building toward a position where it captures value from AI model consumption and AI-initiated commerce — not just from payment processing. The abandoned PayPal bid reinforces this reading: Stripe appears to have chosen depth in AI infrastructure over breadth in consumer payments. Other PSPs should assess whether their own product roadmaps have an answer to this strategic shift.

State and private digital currency infrastructure are developing on converging timelines. The digital euro's legislative progress (2029 target) and Visa's stablecoin platform expansion are not unrelated. Both reflect a structural shift in how settlement assets are conceived — away from bank deposits and card rails as the only options. PSPs and merchants will eventually need to support multiple settlement layers. The question of which layer carries which transaction type, and at what cost, is becoming a near-term commercial decision rather than a distant theoretical one.

What to Watch

• KYA framework adoption by processors and merchants — Watch whether Visa, Mastercard, and Ant publish a formal specification timeline and whether major PSPs outside the founding trio (notably Stripe, PayPal, Adyen) signal alignment or develop competing approaches. Near-term: Q4 2026.

• Issuing bank participation terms in Mastercard Agent Pay — The commercial and liability terms on which banks join Agent Pay will determine whether agentic payments scale or stall. Look for public announcements from top-10 issuing banks in Europe and North America. Timeframe: next two quarters.

• Digital euro merchant pilot participation — The Eurosystem's call for e-commerce and m-commerce merchants is open now. Which merchant categories and PSPs participate will signal where the ECB expects early adoption. Watch for pilot scope announcements before end-2026.

• OpenRouter integration into Stripe's product suite — Stripe has not yet disclosed how OpenRouter's model routing will be packaged for customers. Watch for product announcements that combine payment processing with AI token optimisation, and for pricing models that could disrupt existing AI infrastructure vendors.

• Visa-Mastercard interchange settlement outcome — Merchant objections to the proposed fee settlement are active. A rejection or renegotiation would have material implications for interchange economics across the industry. No confirmed timeline in the available material.

• Regulatory guidance on agent accountability — The KYA framework references a SAFR regulatory framework, but no jurisdiction has yet published binding rules on who is liable when an AI agent initiates a fraudulent or erroneous payment. Watch for guidance from EU, UK FCA, and MAS given the Singapore geography of the Ant-Visa-Mastercard announcement.

Coverage Note

The database contained nine developments. These were merged into five stories: D1 stood alone as the KYA framework; D2 and D3 were merged as overlapping coverage of live agentic payment deployments; D5 and D9 were merged as duplicate coverage of the Stripe-OpenRouter acquisition; D6 stood alone as the digital euro regulatory development; and D4, D7, and D8 were merged as overlapping coverage of Visa's multi-front expansion. No entries were excluded — all described events that occurred in the period.

Evidence quality is mixed. The Stripe acquisition (D5/D9) has the strongest evidence base: eight sources including tier-1, confidence 78–82. The digital euro item (D6) has ten sources of which eight are tier-1, but the 2029 issuance date is explicitly flagged as possibly conditional. The agentic payments items (D1, D2, D3) carry zero tier-1 sources across all entries, which is a material weakness for stories of this significance — treat specific product claims and partnership details as unverified until confirmed by primary sources. The Visa multi-front item (D7) carries a confidence score of 45 on the merchant settlement component; that sub-item should be treated as low confidence. Readers should seek primary source confirmation before acting on any item with zero tier-1 sourcing.

The Week in One Paragraph

Three storylines dominated Week 38. Stripe made its most consequential strategic move in years, acquiring AI model gateway OpenRouter for $7.5 billion after abandoning a $53 billion joint bid for PayPal with Advent International — a pivot that signals where Stripe believes the next layer of financial infrastructure sits. Simultaneously, Visa, Mastercard, and Ant International formalised a Know-Your-Agent framework in Singapore, with live proof-of-concept transactions already settled on Visa rails, marking the moment agentic commerce moved from whitepaper to working infrastructure. In Europe, the digital euro advanced on two fronts: the European Parliament approved the enabling Regulation, and the ECB moved into delivery mode with a 36-firm pilot including Revolut, an accessibility-first app, and a tokenised asset ecosystem vision — though issuance remains a 2029 proposition at earliest. The common thread is that payments infrastructure is being rebuilt around AI agents and programmable money simultaneously, and the decisions made in standards bodies and M&A rooms this week will shape the competitive landscape for a decade.

Top Developments

1. Stripe acquires OpenRouter for $7.5 billion and walks away from PayPal

What happened: Stripe and Advent International abandoned their joint $53 billion bid for PayPal, which had been tabled at $60.50 per share in July 2026. Separately, Stripe agreed to acquire OpenRouter — an AI model marketplace routing requests across more than 400 models from over 80 providers — for approximately $7.5 billion, with $1.5 billion allocated to founders.

Why it matters: The two moves together define Stripe's strategic direction: away from acquiring legacy consumer payments volume and toward owning the economic infrastructure layer for AI companies. OpenRouter's routing and optimisation function for AI models is structurally analogous to what payment orchestration does for transactions. If AI inference becomes a high-volume, margin-sensitive workflow, controlling the routing layer is a durable position. For PayPal, the abandoned bid removes a near-term catalyst and increases pressure on management to articulate a standalone value creation path — analysts have flagged a Venmo spin-off or share buybacks as likely responses.

Companies: Stripe, Advent International, PayPal Holdings Inc., OpenRouter

Impact: 68 | Confidence: 82

2. Visa, Mastercard, and Ant International launch Know-Your-Agent framework; live agentic transactions settle on Visa rails

What happened: Ant International, Visa, and Mastercard announced a collaboration on 10 September 2026 to develop an interoperable Know-Your-Agent (KYA) framework, combining Visa's Trusted Agent Protocol, Mastercard's Verifiable Intent, and Ant International's Agentic Mobile Protocol, with initial focus on Singapore. Separately, Nuvei completed a live proof of concept with Visa, Arvato Systems, and Kings and Priests in July 2026, demonstrating merchant AI agents executing purchases on behalf of shoppers with settlement on live Visa rails. Mastercard introduced Agentic Tokens and Visa launched an Intelligent Commerce framework, both built on tokenisation.

Why it matters: The KYA framework is the agentic equivalent of EMV — a shared identity and trust standard without which interoperable machine-initiated payments cannot scale. The fact that a live proof of concept has already settled on production rails means this is no longer a research exercise. For acquirers and processors, the immediate challenge is fraud prevention: existing models are calibrated for human-initiated transaction patterns. For merchants, the question is whether agent-initiated purchases will carry different liability, fee, or chargeback treatment. Consumer trust remains the stated adoption barrier, but the infrastructure constraint is now closer to resolution than the demand constraint.

Companies: Ant International, Visa, Mastercard, Nuvei, Arvato Systems, Kings and Priests, Accertify, ReconArt, Yuno, Paystand

Impact: 72 | Confidence: 65

3. European Parliament approves digital euro Regulation; ECB enters delivery phase with 36-firm pilot

What happened: The European Parliament has approved the digital euro Regulation. The ECB has selected 36 firms, including Revolut, for a pilot, and is developing a scheme rulebook with market participants. The ECB is simultaneously building an accessibility-first digital euro app and articulating a vision for a tokenised European financial ecosystem with central bank money at its core. Mandatory acceptance and legal tender status are established in the legislative framework, with cost controls on merchant fees. The ECB indicates issuance could occur in 2029, subject to a Governing Council decision once the Regulation is formally adopted.

Why it matters: Mandatory acceptance with fee controls is the provision that matters most to merchants and PSPs. The combination of legal tender status and cost caps structurally resembles the interchange regulation playbook — it removes merchant optionality while constraining the revenue model for intermediaries. The 2029 issuance horizon is long, but the rulebook and pilot work happening now will determine which PSPs and banks are positioned as distribution partners versus disintermediated. Revolut's inclusion in the pilot is notable given its scale across the euro area. The parallel data point — cash acceptance at 92% and mobile payment acceptance rising from 36% to 68% — suggests the ECB is entering a market already in rapid transition.

Companies: European Central Bank, European Commission, Council of the European Union, Revolut

Impact: 78 | Confidence: 72

Note: Confidence on the parliamentary approval specifically is 55 in the source material; the formal adoption date is not confirmed.

4. Visa expands across blockchain lending, stablecoin cards, and embedded cross-border payments

What happened: Visa reported 200% year-over-year growth in stablecoin-linked card programs, with more than 160 active programs, and is expanding data offerings to blockchain lenders including a pilot with Credit Coop for onchain credit. Visa also enabled embedded international money transfers via PayMitto for community banks, with Texas First Bank launching a cross-border P2P product powered by Visa Direct. A regulatory notification was issued to Australian travel management companies regarding RBA surcharge prohibition changes affecting the Visa network from 1 October 2026.

Why it matters: The stablecoin card growth figure, if accurate, indicates that the crypto-to-fiat bridge is becoming a mainstream acquiring category rather than a niche. Visa's move to supply data to blockchain lenders extends its network value proposition beyond transaction rails into credit underwriting infrastructure — a meaningful expansion of the network's role. The community bank embedded payments play via PayMitto illustrates how Visa Direct is being used to retain smaller institutions that might otherwise route around card rails entirely.

Companies: Visa, MoonPay, PayMitto, Credit Coop, Transcard, Texas First Bank

Impact: 62 | Confidence: 55 | low confidence

5. Agentic finance tooling reaches enterprise availability across reconciliation, B2B, and remittance

What happened: ReconArt integrated with the Solana blockchain to support reconciliation of stablecoin and AI-initiated agentic payment transactions. Paystand launched general availability of its agentic finance suite, covering accounts receivable, spend management, and reporting on its B2B payment network. Yuno raised a $45 million Series B to fund its agentic AI strategy and expansion into Gulf markets.

Why it matters: Reconciliation has historically been the last workflow to adapt to new payment types — its arrival signals that agentic and stablecoin payment volumes are reaching a threshold where enterprise finance teams need systematic tooling, not manual workarounds. Paystand's GA launch and Yuno's funding round indicate that B2B is the near-term commercial beachhead for agentic payments, ahead of consumer-facing use cases where trust barriers remain higher.

Companies: ReconArt, Paystand, Yuno

Impact: 62 | Confidence: 45 | low confidence

Themes This Week

The infrastructure layer is being contested simultaneously across AI and programmable money. Stripe's OpenRouter acquisition, the KYA framework, and the digital euro pilot are not separate stories. Each represents a different actor staking a claim on what the next generation of payment infrastructure looks like — AI routing rails, agent identity standards, and central bank digital settlement respectively. The decisions made in standards bodies and boardrooms this week will determine which layer captures value in a world where transactions are increasingly machine-initiated and settled in programmable currency.

Mandatory standards are arriving before the market is ready. The digital euro's mandatory acceptance requirement and the KYA framework's push for interoperability both reflect a pattern: regulators and dominant networks are setting standards ahead of proven consumer demand. Cash acceptance at 92% and consumer trust barriers in agentic commerce both suggest the demand side is lagging. PSPs and merchants face the cost of compliance and integration before the revenue case is clear.

Agentic commerce is moving from concept to infrastructure in a single quarter. A live proof of concept on production Visa rails in July, a formal interoperability framework in September, and enterprise reconciliation tooling reaching general availability in the same week mark a compression of the typical hype-to-infrastructure cycle. The fraud and liability frameworks have not kept pace — this is the near-term operational risk for any acquirer or processor that enables agent-initiated transactions.

Stripe's strategic pivot reframes the competitive map. By abandoning PayPal and acquiring OpenRouter, Stripe has signalled that it does not intend to compete for legacy consumer payments volume. This repositions the competitive dynamic: PayPal is left without a near-term acquirer and under pressure to restructure, while Stripe moves toward AI infrastructure where its developer-first positioning is a genuine advantage. Other processors will need to decide whether to follow into AI infrastructure or double down on payments volume.

What to Watch

• Digital euro formal adoption date and Governing Council issuance decision — the parliamentary approval is confirmed but the formal adoption trigger for the Governing Council's issuance decision is not yet set. Watch for the Council of the EU's formal adoption step, which starts the clock toward a potential 2029 launch.

• KYA framework adoption beyond Singapore — the initial focus is Singapore. Whether Visa, Mastercard, and Ant International extend the framework to the EU and US, and whether other processors adopt it or develop competing standards, will determine whether KYA becomes the industry baseline or one of several fragmented approaches.

• PayPal's strategic response to the abandoned bid — with Stripe and Advent International out, PayPal faces valuation pressure. Watch for announcements on Venmo structure, share buyback programmes, or alternative transaction discussions before end of Q4 2026.

• Stripe-OpenRouter integration timeline and commercial terms — the acquisition is announced but integration details are not disclosed. The pricing model for AI routing services will indicate whether Stripe intends to monetise this as a standalone product or bundle it into existing payment infrastructure.

• Fraud and liability framework for agent-initiated transactions — no regulator or network has published definitive rules on chargeback liability or fraud responsibility for agentic purchases. Watch for guidance from Visa and Mastercard alongside the KYA rollout, and for regulatory responses in the EU and US.

• Merchant fee structure in final digital euro legislation — cost controls are established in principle but the specific fee caps are not yet confirmed in the material. The final merchant fee structure will determine whether the digital euro is a cost-reduction opportunity or a compliance cost for merchants.

Coverage Note

The database contained 15 developments. After review:

Merged: D1 and D3 describe the same ECB digital euro delivery story from different source runs and were merged with D9 (parliamentary approval and pilot) into a single item, using the highest impact score of 78. D2, D4, D5, D6, and D12 all cover the agentic commerce infrastructure story from different angles and source sets; these were merged into one item (KYA framework and live PoC) and a separate supporting item on enterprise tooling (D14). D10, D11, and D15 cover overlapping Visa partnership activity and were consolidated. D7 and D13 both cover the Stripe/PayPal abandonment; D8 covers the OpenRouter acquisition — all three were merged into a single Stripe item.

Excluded: None excluded on the grounds of being non-events; all entries described something that happened. However, D14 was retained as a lower-confidence supporting item rather than a standalone development given its confidence score of 45.

Evidence quality: Mixed. The digital euro and Stripe/OpenRouter items have reasonable source depth (up to 8 sources, including tier-1). The agentic commerce cluster carries zero tier-1 sources across all entries, which is a material weakness — the KYA framework announcement and the live PoC details should be treated as directionally credible but not confirmed. The Visa multi-partnership item has low confidence (55) and no tier-1 sources. Readers should weight the agentic and Visa items accordingly.

The Week in One Paragraph

Three structural forces dominated Week 37: the race to build autonomous payment infrastructure, the consolidation of stablecoin settlement as a mainstream network capability, and the ECB's concrete steps toward a digital euro. Agentic payments moved from conference-circuit concept to live deployments and significant M&A, with Stripe's reported $7 billion acquisition of OpenRouter and a live Visa proof-of-concept on European rails signalling that the largest players are committing capital, not just roadmaps. Visa simultaneously broadened its stablecoin settlement footprint across Asia and globally through a cluster of partnerships, contrasting sharply with Mastercard's acquisition-led approach via BVNK. The ECB named Worldline as a pilot provider and published accessibility standards for the digital euro app, moving the project from policy document to procurement reality. Against this backdrop, the Stripe-Advent pursuit of PayPal remains unresolved, with conflicting reports on whether the $53 billion deal is dead or merely paused — a material uncertainty for the structure of the US payments market.

Top Developments

1. Agentic Payments Cross from Experiment to Infrastructure Investment

What happened: Multiple players made concrete commitments to autonomous AI-driven payment execution in the same week. Stripe reportedly acquired AI infrastructure platform OpenRouter for approximately $7 billion. Visa completed a live proof of concept with Nuvei and Arvato Systems for agentic commerce payments across European issuers. XDC Network demonstrated Claude executing a payment via the x402 machine-payments standard and USDC. ReconArt integrated Solana for reconciliation of AI-initiated transactions. Yuno raised $45 million Series B with an explicit agentic AI strategy.

Why it matters: The payments stack is being restructured around a new principal: the AI agent rather than the human cardholder. This disrupts assumptions embedded in fraud models, authorisation frameworks, liability rules, and interchange economics. The complexity is not in moving money — existing rails handle that — but in integrating authorisation, delegation, reconciliation, and exception handling into workflows where no human is present at the moment of transaction. Whoever owns the authorisation and routing layer for agent-initiated payments owns a structurally advantaged position. Stripe's OpenRouter acquisition is a direct bid for that layer.

Companies: Stripe, OpenRouter, Visa, Nuvei, Arvato Systems, XDC Network, ReconArt, Solana, Circle, Anthropic, Yuno, Mastercard, Accertify, Bluefin, Peakflo, DEUNA, PayPal, Affirm, Walmart

Impact: 72 | Confidence: 62 | low confidence on several individual claims; no tier-1 sources across this cluster

2. Visa Builds Open Stablecoin Settlement Network Across Asia and Globally

What happened: In August 2026, Visa announced a cluster of stablecoin settlement initiatives: joining Singapore's MAS-supervised BLOOM framework with Nium as pilot partner; signing strategic partnerships with South Korean firms Dunamu and Shinhan Financial; integrating stablecoin settlement into Visa Direct across 195 countries through partnerships with ZeroHash and Lightspark; and reporting $7 billion in annualised transaction volume on its stablecoin pilot as of April 2026. Visa supports multiple stablecoins including USDC, RLUSD, and Open USD across nine blockchain networks.

Why it matters: Visa is positioning itself as the open-network aggregator for stablecoin settlement — supporting multiple assets and chains rather than owning one. This contrasts directly with Mastercard's $1.8 billion acquisition of BVNK, which bets on vertical integration. For PSPs and banks, Visa's approach lowers the barrier to stablecoin settlement without requiring a new counterparty relationship. The BLOOM pilot under MAS supervision is significant: regulatory cover from a credible central bank gives institutional clients the compliance anchor they need to proceed.

Companies: Visa, ZeroHash, Lightspark, Circle Internet Group, Monetary Authority of Singapore, Nium, Dunamu, Shinhan Financial, Mastercard, Nuvei, Bluefin, Featurespace, Inflowpay, Pathward, TabaPay, Fransabank Lebanon

Impact: 68 | Confidence: 65 | no tier-1 sources across this cluster

3. ECB Moves Digital Euro from Policy to Procurement

What happened: The ECB selected Worldline as a technical and acquiring service provider for the Eurosystem's digital euro pilot programme. Separately, the ECB published accessibility standards for the digital euro app that exceed European Accessibility Act requirements, and ECB leadership outlined tokenisation and DLT as central to a new European digital finance ecosystem with central bank money at its core. Current acceptance data shows cash at 92%, card at 88%, and mobile payments rising sharply from 36% to 68% between 2024 and 2026.

Why it matters: Worldline's selection is the first concrete procurement signal that the digital euro has a delivery timeline, not just a design. For acquirers and PSPs operating in the euro area, this raises immediate questions about integration requirements, interchange economics under a CBDC model, and whether the ECB's accessibility-first design principles will constrain commercial product differentiation. The mobile payment acceptance surge — nearly doubling in two years — indicates the consumer infrastructure is maturing faster than the regulatory framework.

Companies: European Central Bank, Worldline

Impact: 72 | Confidence: 82 | strong; 3 tier-1 sources; D1 and D3 merged as same story

4. Stripe-PayPal Deal: Abandoned, Then Reportedly Revived

What happened: Stripe and Advent International walked away from a joint acquisition bid for PayPal after PayPal rejected a $60.50-per-share offer valued at over $53 billion. Subsequent reporting indicates PayPal is now actively negotiating a sale, with the Stripe-Advent consortium potentially re-engaged. Antitrust risk is the central regulatory concern given overlapping merchant processing operations.

Why it matters: A Stripe-PayPal combination would create a payments entity with dominant positions across merchant acquiring, consumer wallets, BNPL access, and developer infrastructure simultaneously. Regulators would likely require divestitures — Venmo and Braintree are the most-cited candidates. Even if the deal does not close, the pursuit signals that Stripe views scale in consumer-facing payments as strategically necessary, not optional. PayPal's willingness to negotiate a sale reflects the pressure its standalone valuation is under.

Companies: Stripe, PayPal Holdings Inc., Advent International

Impact: 72 | Confidence: 55 | low confidence; no tier-1 sources; conflicting reports on deal status

5. Adyen Deepens Embedded Finance Offering Through Epos Now Partnership

What happened: Adyen announced an integrated embedded finance partnership with SaaS platform Epos Now, combining payments, business accounts, capital, and card issuing into a single platform. Issuing and Accounts are available in the EU, UK, and US; Capital is also available in Australia, Canada, and Spain.

Why it matters: This is a relatively contained but strategically consistent move. Adyen's embedded finance model — acting as both gateway and acquiring bank — allows SaaS platforms to monetise financial services without building their own regulated infrastructure. For SMB-focused SaaS providers, this is a direct competitive pressure: platforms that do not embed financial services risk losing merchants to those that do. The multi-market rollout signals Adyen is scaling this model beyond its initial EU base.

Companies: Adyen, Epos Now

Impact: 58 | Confidence: 72

Themes This Week

Autonomous agents as a new payment principal. D2, D5, D6, D9, D11, and D12 all point the same direction: the industry is preparing for a world where AI agents, not humans, initiate and authorise transactions. The infrastructure gaps are not in rails but in delegation frameworks, fraud models calibrated for non-human behaviour, and reconciliation systems that can handle AI-initiated flows. Stripe's OpenRouter acquisition and Visa's live proof of concept are the two most concrete signals that this is moving from roadmap to capital allocation.

Stablecoin settlement as network infrastructure, not niche product. Visa's cluster of partnerships — MAS-supervised, multi-chain, multi-stablecoin — and the $7 billion annualised volume figure indicate that stablecoin settlement is being absorbed into mainstream network infrastructure. The strategic divergence between Visa (open aggregation) and Mastercard (acquisition of BVNK) will be a defining competitive dynamic over the next 18 months. PSPs and banks choosing settlement partners now are making a structural bet on which model prevails.

Regulatory anchoring as a competitive asset. The ECB's digital euro pilot and the MAS-supervised BLOOM framework both illustrate the same dynamic: central bank or regulatory involvement is becoming a prerequisite for institutional adoption of new payment infrastructure. Worldline's selection by the ECB and Visa's participation in BLOOM are not just commercial wins — they are regulatory positioning that competitors will find difficult to replicate quickly.

What to Watch

• Stripe-PayPal deal status — Whether the Stripe-Advent consortium formally re-engages or PayPal pursues alternative transactions (Venmo spin-off, buyback programme). Conflicting reports make this the highest-uncertainty item in the week's news.

• Digital euro pilot scope — Which additional providers join Worldline in the Eurosystem pilot, and what the ECB publishes on interoperability standards for tokenised asset platforms. The procurement signal is new; the timeline remains opaque.

• BLOOM pilot outcomes — Whether the MAS-supervised framework expands beyond Nium and what transaction volumes and settlement performance data emerge. This is the most credible live test of institutional stablecoin settlement under regulatory supervision.

• Agentic commerce regulatory response — Whether any regulator (EU, US, Singapore) issues guidance on autonomous AI payment execution, delegation of authority, or liability allocation for agent-initiated transactions. The x402 standard and Visa's proof of concept are ahead of any regulatory framework.

• Stripe OpenRouter integration — Completion of the reported $7 billion acquisition and the first product announcements integrating OpenRouter's AI routing capabilities into Stripe's payment infrastructure. Watch for customer adoption metrics.

• Dunamu and Shinhan Financial product launches — Regulatory approval timeline in South Korea for Visa-partnered stablecoin payment and remittance products. South Korea's regulatory environment for crypto-linked financial products will be the binding constraint.

Coverage Note

The database contained 17 developments. Significant clustering was required: D1 and D3 are the same ECB digital euro story from different source runs and were merged into a single item (Impact 72, Confidence 82). D2, D5, D6, D9, and D11 all cover agentic payments from overlapping angles and were merged into one item. D7, D8, D12, D13, and D17 all cover Visa's stablecoin and partnership activity and were merged into two items (stablecoin settlement and fraud/security). D4 and D10 cover the same Stripe-PayPal story and were merged. D14 and D15 overlap substantially on Stripe-OpenRouter and DBS and were merged into the Stripe-PayPal item with the OpenRouter detail incorporated into the agentic payments item.

No entries were excluded on the grounds of being non-events; all contained at least one identifiable development.

Evidence quality is weak overall. Eleven of seventeen entries carried zero tier-1 sources. The agentic payments cluster in particular rests entirely on non-tier-1 sourcing, and confidence scores in the 52–65 range reflect genuine uncertainty about which specific product launches and partnerships are confirmed versus reported. The Stripe-PayPal item carries an explicit low-confidence flag. The ECB digital euro item is the strongest in the database, with three tier-1 sources and high internal consistency. Readers should treat agentic payments claims as directionally credible but not individually verified.

The Week in One Paragraph

The digital euro moved from preparation into active pilot testing, with Worldline confirmed as an acquiring and technical partner, 36 PSPs selected for a 2027 pilot, and retail availability targeted for 2029 — the most concrete implementation timeline the ECB has published. Simultaneously, Visa and Mastercard are racing to own the infrastructure layer for two converging technology shifts: stablecoin settlement and agentic commerce. Visa is pursuing a partnership-led stablecoin strategy across Asia and globally, while Mastercard acquired BVNK; both networks have now launched tokenization frameworks specifically designed for AI agent-initiated payments. Stripe's reported $7 billion acquisition of OpenRouter and its DBS partnership signal that PSPs are also repositioning for an AI-native payments stack. Taken together, the week's developments describe a payments industry simultaneously managing a decade-long public infrastructure project in Europe and a much faster-moving private-sector race to define the rails for autonomous commerce.

Top Developments

1. Digital Euro Enters Pilot Phase with Confirmed Partners, 2027–2029 Rollout Timeline

What happened: The ECB has transitioned the digital euro from preparation into active pilot testing, with approximately 70 market participants — banks, fintechs, merchants, and PSPs — involved in an innovation platform. Worldline has been confirmed as both acquiring PSP and technical service provider. A formal pilot involving 36 PSPs is scheduled for 2027, with retail availability targeted for 2029. Legislation covering mandatory merchant acceptance and fee controls is under autumn 2026 negotiation.

Why it matters: Mandatory acceptance provisions, if enacted, will force every euro-area acquirer and merchant to support a new payment instrument with fee structures set by regulators rather than the market. The 2029 retail launch is far enough away to seem abstract, but the 2027 PSP pilot and the autumn 2026 legislative window are near-term forcing functions. PSPs not engaged in the sandbox now risk being late to shape integration standards. The ECB's explicit positioning of the digital euro as a stablecoin alternative also has direct implications for MiCA-compliant euro stablecoin issuers.

Companies: European Central Bank, European Commission, Eurosystem central banks, Worldline, Adyen, Revolut, SumUp.

Impact: 78 · Confidence: 75

2. Visa and Mastercard Launch Competing Stablecoin Settlement Strategies at Scale

What happened: Visa reached $7 billion in annualized stablecoin transaction volume as of April 2026 and in August 2026 expanded through partnerships with ZeroHash, Lightspark, Nium (via the MAS-supervised BLOOM framework), Dunamu, and Shinhan Financial Group, supporting USDC, RLUSD, and Open USD across nine blockchain networks. Mastercard took the acquisition route, buying BVNK for up to $1.8 billion and launching an 85-partner crypto programme. European banks formed the Qivalis consortium — 37 members — to issue MiCA-compliant euro stablecoins.

Why it matters: Stablecoins are no longer a peripheral crypto product; they are being embedded in merchant acquiring, cross-border settlement, payroll, and corporate treasury. The divergence in strategy — Visa's open-network partnership model versus Mastercard's vertical acquisition — will determine which network captures more of the settlement margin as volumes scale. For PSPs and acquirers, the question is no longer whether to support stablecoin rails but which network's infrastructure to build on first. The Qivalis consortium adds a European institutional dimension that intersects directly with the digital euro legislative debate.

Companies: Visa, Mastercard, BVNK, Circle Internet Group, ZeroHash, Lightspark, Nium, Dunamu, Shinhan Financial Group, Qivalis, Monetary Authority of Singapore.

Impact: 72 · Confidence: 68

3. Agentic Commerce Infrastructure Moves to Live Production on Card Rails

What happened: Nuvei completed a live proof-of-concept with Visa, Arvato Systems, and Kings and Priests in July 2026, demonstrating merchant AI agents initiating purchases with granular spend controls and network tokenization on live Visa rails. Mastercard launched Agentic Tokens under its Agent Pay programme; Visa deployed its Intelligent Commerce framework. Google and OpenAI have published competing open standards — Universal Commerce Protocol and Agentic Commerce Protocol respectively. PPRO partnered with BLIK to enable AI payments in Poland.

Why it matters: The move from proof-of-concept to live rails is the critical threshold. Network tokenization is emerging as the security layer of record for agent-initiated transactions, which means acquirers and issuers that have not fully implemented network tokenization are structurally behind. The competing open standards from Google and OpenAI introduce a protocol fragmentation risk that mirrors the early days of mobile payments. Fraud liability frameworks for autonomous transactions remain unresolved, which is the single largest barrier to enterprise scale-up.

Companies: Visa, Mastercard, Nuvei, Arvato Systems, Kings and Priests, Google, OpenAI, PPRO, BLIK, Thredd, ACI Worldwide, Accertify.

Impact: 72 · Confidence: 60 · low confidence on Morgan Stanley projections and protocol adoption claims (zero tier-1 sources across underlying entries)

4. Stripe Acquires OpenRouter and Partners with DBS, Signalling AI-Native PSP Repositioning

What happened: Stripe is reportedly acquiring AI infrastructure platform OpenRouter for approximately $7 billion, gaining access to its customer base and AI routing capabilities. On 27 August 2026, Stripe announced a strategic partnership with DBS to accelerate cross-border payments and agentic commerce in Asia, combining DBS's pan-Asia banking connectivity with Stripe's programmable payments platform.

Why it matters: A $7 billion AI infrastructure acquisition by a PSP is a category-defining signal: Stripe is betting that payment routing will become an AI function, not a configuration function. The DBS partnership gives Stripe regulated banking infrastructure across Asia without requiring its own banking licences — a model other large PSPs will evaluate. Combined with the OpenRouter acquisition, Stripe is positioning to serve both the human-initiated and agent-initiated payment stacks simultaneously.

Companies: Stripe, OpenRouter, DBS, DBS Bank.

Impact: 62 · Confidence: 55 · low confidence (acquisition reported, not confirmed; zero tier-1 sources)

5. Apple Pay Launches in Philippines; JPMorgan Chase Takes Over Apple Card Issuance

What happened: Apple launched Apple Pay in the Philippines on 4 August 2026, partnering with Chinabank, GoTyme Bank, Metrobank, and UnionBank, with acceptance across 5,000-plus online merchants via Maya Business. Concurrently, JPMorgan Chase is replacing Goldman Sachs as Apple Card issuer, with the transition expected to complete by 2028.

Why it matters: The Philippines launch extends Apple Pay into a high-growth Southeast Asian market where digital wallet penetration is accelerating. The issuer transition on Apple Card is more structurally significant: JPMorgan Chase absorbing the programme signals that large universal banks — not fintechs — are the natural long-term home for co-brand credit infrastructure at scale. Goldman's exit is a data point for any bank evaluating consumer fintech partnerships against core return requirements.

Companies: Apple Inc., JPMorgan Chase, Maya Business, Chinabank, GoTyme Bank, Metrobank, UnionBank.

Impact: 62 · Confidence: 65

Themes This Week

The infrastructure layer is being contested on three fronts simultaneously. Stablecoins, agentic commerce, and the digital euro are each demanding new settlement, tokenization, and authentication infrastructure. What is striking is that the same networks — Visa, Mastercard — are building for all three simultaneously, while PSPs like Stripe and Nuvei are positioning as the orchestration layer above them. The risk for mid-tier acquirers and regional PSPs is being squeezed between network-level infrastructure and AI-native orchestrators.

Tokenization is becoming the universal security primitive. Network tokenization appears across agentic commerce (Visa/Nuvei proof-of-concept), digital wallet expansion (Visa Digital Wallet Enabler APIs), and stablecoin-to-card conversion flows. The 4.6% authorization rate lift cited for tokenized card-not-present transactions gives acquirers a commercial argument to accelerate adoption independent of the agentic commerce narrative. Issuers and acquirers that treat tokenization as a compliance checkbox rather than a revenue lever are misreading the market.

Public and private digital money infrastructure are on a collision course in Europe. The ECB's explicit framing of the digital euro as a stablecoin alternative, combined with the Qivalis consortium's MiCA-compliant euro stablecoin issuance and the PSD3/PSR finalisation, means European PSPs face three overlapping regulatory and commercial frameworks for digital money — all maturing between 2026 and 2029. Mandatory acceptance rules for the digital euro, if enacted, will force product and compliance decisions well before the 2029 retail launch.

AI agent payments are ahead of their fraud and liability frameworks. Multiple developments this week describe live agentic payment transactions, but the fraud prevention and regulatory liability questions remain open across all of them. Accertify, ACI Worldwide, and The Clearing House are referenced in the context of fraud infrastructure, but no resolved framework for autonomous transaction liability has been published. This gap will constrain enterprise adoption more than any technical barrier.

What to Watch

• Autumn 2026: EU legislative negotiations on digital euro mandatory acceptance and fee control rules — the outcome will set the commercial terms for every euro-area acquirer and merchant.
• 2027: Selection criteria and performance metrics for the 36-PSP digital euro pilot — which PSPs are included will signal which integration architectures the ECB is standardising around.
• Near-term: Visa's announcement of its new stablecoin settlement partner (referenced as pending in the source material) — this will clarify whether Visa's open-network model extends to a primary settlement relationship or remains a multi-partner hedge.
• Near-term: Completion and regulatory approval of Stripe's OpenRouter acquisition — if confirmed at $7 billion, it will trigger reassessment of AI infrastructure valuations across the PSP sector.
• H1 2027: The Clearing House tokenized-deposit network launch — a US institutional counterpart to the European stablecoin and digital euro developments, with direct implications for cross-border settlement.
• Ongoing: Regulatory guidance on liability for autonomous agent-initiated transactions — the absence of a framework is currently the binding constraint on enterprise agentic commerce deployment.

Coverage Note

The database contained 25 developments. Seven entries (D1, D2, D3, D4, D9, D10, and partially D7) were merged into a single digital euro item — they describe the same ECB programme from different source runs with overlapping but non-contradictory detail; the highest impact score (78) was retained. Five entries covering Visa stablecoin strategy (D11, D12, D15, D16, D18) were consolidated with the Mastercard stablecoin entry (D6) into one item. Five agentic commerce entries (D5, D13, D14, D17, D22) were merged. D8 (PSD3 finalisation) was noted as contextually relevant but carries a confidence score of 48 — below the threshold for standalone treatment — and was folded into thematic commentary rather than a top development. D7 (3DS/SCA evolution) contained no discrete event in this period and was excluded as a market-state description rather than a development. D23 (biometric authentication) similarly described market trends without a specific datable event and was excluded.

Overall evidence quality is moderate. The digital euro cluster has the strongest sourcing (up to 5 tier-1 sources). The agentic commerce and Stripe acquisition clusters have zero tier-1 sources, which is a material weakness — the Stripe/OpenRouter figure and the Morgan Stanley projections should be treated as indicative rather than confirmed. Confidence scores below 55 are flagged inline.

The Week in One Paragraph

The dominant story this week is the digital euro's progression from preparation into active pilot testing, with the ECB publishing innovation platform results and confirming a legislative and rollout timeline stretching to 2029. That timeline is long, but the mandatory acceptance provisions and fee controls embedded in draft legislation make this a structural concern for every merchant acquirer and PSP operating in the eurozone. Alongside it, agentic payments moved from concept to live proof of concept, with Visa and Mastercard both advancing framework standards and Nuvei completing a transaction on Visa rails with European issuers. The two stories are connected: the infrastructure layer of payments — who sets the rails, who controls the token, who authorises the agent — is being contested simultaneously by central banks and by card networks. Adyen's Toast partnership expansion is a real commercial development but operates at a different order of magnitude.

Top Developments

1. ECB Moves Digital Euro into Pilot Phase and Confirms 2029 Retail Launch

What happened: The ECB has transitioned the digital euro from preparation into active pilot testing, publishing results from an innovation platform involving approximately 70 market participants including banks, fintechs, merchants, and PSPs. Key capabilities tested include offline payment functionality, conditional payments, and advanced accessibility features. A formal pilot involving 36 payment service providers is scheduled for 2027, with retail availability targeted for 2029.

Why it matters: The mandatory acceptance provisions in draft legislation mean merchants and acquirers cannot treat this as optional. Fee controls embedded in the legislative proposals will constrain the commercial model for PSPs handling digital euro transactions. The explicit positioning of the digital euro as a substitute for stablecoins signals regulatory intent to shape the competitive landscape of digital assets in Europe. PSPs and acquirers have a narrow window — roughly 2027 to 2029 — to build integration capability before retail rollout.

Companies: European Central Bank, European Commission, Eurosystem central banks.

Impact: 78 · Confidence: 75 (three overlapping entries with partially independent sourcing; legislative details carry moderate confidence given autumn 2026 negotiations are ongoing)

2. Agentic Payments Moves from Framework to Live Proof of Concept

What happened: Visa's Intelligent Commerce and Mastercard's Agent Pay frameworks are being standardised as infrastructure for AI agents to initiate and execute payments autonomously. Nuvei completed a live proof of concept on Visa rails involving Arvato Systems and Kings and Priests, with multiple European issuers participating. Use cases span both issuer-side workflows — reconciliation, onboarding, dispute handling — and merchant-side autonomous transaction initiation.

Why it matters: A live proof of concept on card rails with named counterparties is a meaningful step beyond white-paper positioning. The open question is liability: when an autonomous agent initiates a transaction, existing authorisation and chargeback frameworks were not designed for that principal structure. Regulatory guidance is absent. PSPs and issuers that wait for clarity risk being excluded from early commercial deployments; those that move now carry unpriced liability exposure.

Companies: Visa, Mastercard, Nuvei, Thredd, ACI Worldwide.

Impact: 68 · Confidence: 55 · low confidence (three sources, none tier-1; proof-of-concept details are thin)

3. Visa Embeds Dispute and Monitoring Infrastructure Across Acquiring Platforms

What happened: Visa's Rapid Dispute Resolution tool is being integrated into merchant platforms including payabl., while the Visa Acquirer Monitoring Programme consolidates fraud and dispute monitoring under tightening thresholds. Concurrently, card processors Marqeta and Lithic are expanding from issuing into merchant acquiring and non-card rails.

Why it matters: VAMP threshold tightening has direct compliance cost implications for acquirers and their merchant portfolios. The simultaneous expansion of Marqeta and Lithic into acquiring compresses the traditional boundary between issuing and acquiring — a structural shift that affects how PSPs position their stack. Visa is becoming embedded as compliance infrastructure, not just a transaction rail, which increases switching costs across the value chain.

Companies: Visa, Marqeta, Lithic, Nuvei, payabl.

Impact: 62 · Confidence: 55 · low confidence (three sources, none tier-1)

4. Adyen Expands U.S. Footprint Through Toast Partnership

What happened: Adyen announced an expansion of its existing partnership with Toast, integrating Adyen's payment processing into Toast's U.S. payments ecosystem across online, in-app, and in-store channels. The expansion builds on an existing international partnership and extends Adyen's reach into the restaurant and hospitality vertical in the United States.

Why it matters: The restaurant and hospitality vertical is a competitive battleground for PSPs offering embedded finance and vertical-specific acquiring. Adyen's move deepens its U.S. enterprise presence in a segment where integrated POS and payments capability is a meaningful differentiator. The commercial significance is real but limited in scope relative to the structural stories this week.

Companies: Adyen, Toast.

Impact: 35 · Confidence: 45 · low confidence (three sources, none tier-1)

Themes This Week

The infrastructure layer is being contested on two fronts simultaneously. The ECB is building a sovereign payment token with mandatory acceptance; Visa and Mastercard are building agentic commerce frameworks on existing card rails. Both efforts are about who controls the authorisation layer in the next generation of payments. PSPs and acquirers face the unusual position of needing to engage with both tracks at once, with different regulatory, technical, and commercial implications from each.

Liability frameworks are lagging product development. Agentic payments introduce a new authorisation principal — the AI agent — for which chargeback, fraud liability, and dispute resolution rules do not yet exist. The digital euro's offline functionality raises similar questions about settlement finality and fraud recovery outside network connectivity. In both cases, the technical capability is advancing faster than the legal and regulatory infrastructure designed to govern it.

The issuing-acquiring boundary is dissolving. Marqeta and Lithic expanding into acquiring, Adyen deepening vertical integration through Toast, and Visa embedding dispute tools directly into merchant platforms all point in the same direction: the clean separation between card issuance and merchant acceptance is becoming commercially unviable as a standalone model. Full-stack positioning is no longer a differentiator — it is becoming the baseline expectation.

What to Watch

• Autumn 2026: Outcome of EU legislative negotiations on digital euro mandatory acceptance and fee control provisions — any amendment here materially changes the commercial model for acquirers and PSPs.
• 2027: Selection of the 36 PSPs for the digital euro pilot programme — inclusion or exclusion will determine which players shape the integration standards that the rest of the market inherits.
• Ongoing: Visa's VAMP threshold tightening schedule — acquirers should track the specific metrics and timelines, as compliance failures carry programme-level consequences.
• Near-term: Regulatory guidance from any EU or U.S. authority on liability allocation for autonomous agent-initiated transactions — absence of guidance is itself a risk signal for PSPs considering early agentic deployments.
• 2029: Digital euro retail availability — merchant POS integration requirements and fee structures will become concrete well before this date; the 2027 pilot is the real decision point for infrastructure investment.

Coverage Note

The database contained seven entries. Three entries (D1, D2, D3) described the same digital euro story from overlapping source sets and were merged into a single item using the highest impact score (78). One entry (D7) was excluded: it described Adyen's market positioning, founding history, and competitive landscape — a company profile, not a development. Five items remained after merging and exclusion, reduced to four in the final report by treating D6 and D7 as a single Adyen story (D7 excluded; D6 retained).

Evidence quality is mixed. The digital euro cluster has the strongest sourcing (up to eight sources, five tier-1) but the three entries show some inconsistency in detail, suggesting different source sets captured different aspects of the same ECB communications. The agentic payments and Visa infrastructure stories (D4, D5) carry no tier-1 sources and should be treated as directionally indicative rather than confirmed. The Adyen-Toast item (D6) similarly has no tier-1 sourcing and low confidence. Readers should weight the digital euro material heavily and treat the remaining items as leads requiring verification.