Institution Insights and redesigned support flow now live

For customers building on Plaid, staying ahead of issues is central to keeping users connected and confident in your product. The support teams behind these products need an intuitive way to manage cases and issue resolution as well as actionable, personalized insights that let them get ahead of problems before users feel the impact.

That's why we've made two significant improvements to the Plaid Dashboard. We've redesigned the support ticket submission flow to reduce friction and get your team to resolution faster, and we've reimagined institution health observability, turning it from a static dashboard into a personalized, actionable view of what issues are impacting your customer traffic.

Get help & manage cases faster

The redesigned support flow makes it faster to find answers and resolve issues, with a cleaner experience and better functionality.

• Known Issues, disruptions Plaid is already tracking, now surface immediately when you start creating a case. This functionality gives your team actionable root cause and resolution guidance so they can find answers without needing to submit the case. When you do need to file a case, we’ve reduced the number of required fields and categories to streamline submission. 

• Instant Help, the AI-powered assistant (formerly Bill), is available as an optional resource throughout the process, accessible via a persistent button whenever you need it. 

• The case management inbox is also redesigned with filters to sort by Known Issues you're following, standard cases, business critical cases, and by relevant financial institutions, making it easier to prioritize and take action.

Together, these updates give your support and service teams the tools to respond to users more proactively and with greater confidence.

Know exactly what issues are impacting your users

Institution Insights marks a significant upgrade to the previous institution status page in the Plaid Dashboard by bringing real-time visibility into the health of Plaid’s connections with financial institutions specifically for your customer traffic. You can now see success rates and error breakdowns filtered to your traffic, with institutions ranked by their share of your traffic so you can immediately see what needs your attention. 

Here’s what you can do now:

• See real-time institution health status based on your customer traffic: The previous institution status tool showed global institution health metrics that reflect the health of all institutions across all Plaid traffic. Now, you’ll also see health metrics by institutions filtered to your traffic specifically so you can understand what issues are impacting your business the most.

• Breakdown success rates and errors: See how success rates and errors for a specific institution vary by product and time to easily understand the health of a connection.

• See actionable issue details: Click into any error to see root cause, user-facing symptoms, and resolution guidance so you can easily diagnose the issue and create proactive customer communications.

• Search by item-level: Look up any item ID to instantly surface the relevant issue details, making it easier to troubleshoot. 

• Follow issues: Subscribe to issues that matter to you and get updates as they resolve.

Customer support and experience teams are using Institution Insights daily to send proactive user communications during incidents to reduce inbound support volume. Additionally, product owners of the Plaid integration are using this tool to diagnose institution-driven conversion issues and determine mitigation without having to build out custom observability. 

Optimize customer success

We're committed to making the Plaid Dashboard the most effective tool for running your integration and serving your users. Institution Insights and the redesigned support flow are the latest steps toward that goal. 

If you're building on Plaid and want to go deeper on Dashboard tooling and observability, we'd love to connect.

This month's updates help you serve more users and make more confident decisions with their financial data. Broader coverage across payments, institutions, mortgages, credit cards, and investments means fewer gaps when you pull account data — and new modeling and reporting help you reduce losses from fraud and payment returns, approve more good applicants, and more.

Featured products: Transfer, Guaranteed Payments, Investments Move

What’s better:

• 🏦 ACH payments for remittance and gaming businesses: Plaid Transfer and Guaranteed Payments now support remittance and gaming. Remittance eligibility is limited to experienced providers with wallet-funding use cases where wallets are held at a US financial institution, and funding must be kept fully separate from outbound remittance flows.
  

• 📈 Investments Move now supports ACATS across 2,700 institutions: Investments Move now works across approximately 2,700 institutions. This more than doubles the coverage available to customers and their users who wish to move their investment accounts via ACATS, allowing for faster growth of assets under management.

Featured products: Plaid Income

What’s better:

• 📄 Expanded income document support: Document Income now parses I-20 forms and Leave and Earnings Statements (LES), expanding support for additional income document types. With broader document coverage, lenders can verify income for international students and military applicants with fewer documentation gaps.
  

Featured products: Protect, Identity Verification

What’s better:

• 🏷️ Submit fraud labels directly in the Protect dashboard: Fraud labels are an important resource for model performance. Customers can now submit “fraud” and “not fraud” labels directly from the Protect Dashboard.
  
  

• 📍 View a customer’s location on a map: In the dashboard, you’ll also see a map that plots a session's location signals against each other. For example: Document Address, Submitted Address, IP Address, or IP Address (VPN) — and shows the distance between them plus a risk level. This helps visualize outliers for deeper investigation, like an IP address in an unexpected country, or a large distance between submitted location and IP location.

• 🔍 Detect manipulated images of ID documents: Fraudsters alter and edit images of ID documents to try to bypass downstream verification checks. IDV’s integrity check inspects the metadata to flag tampered images from virtual camera software and signs of digital editing or AI generation. This feature is designed to counteract AI-driven fraud and flagged 50+ virtual camera attempts in one month.

We shipped a lot more on IDV recently. See the full roundup here.

Featured products: Layer

What’s better:

• 🔀 Intelligent Account Sorting, now configurable in the Layer dashboard: Layer now offers account sorting logic as a template configuration option. Choose the logic that best supports your use case to determine which account is pre-selected — helping users share the bank account that matters most to your business, improving downstream outcomes while keeping user choice fully intact.

• 🏠 Residential address filtering now available in Layer: You can now filter out PO boxes and/or commercial addresses from Layer via a dashboard toggle. When enabled, users without a residential address on file will not be eligible for Layer, and users who attempt to enter a PO box or commercial address are prompted to submit a valid residential address before continuing. This is especially useful in scenarios where residential-only addresses are a hard requirement.

Featured products: Liabilities, Investments

What’s better:

• 🏦 Get mortgage data for 8 of the 10 largest US servicers with 97–99% fill rates: Plaid now supports 8 of the 10 largest mortgage servicers in the US—Rocket Mortgage, Chase, Wells Fargo, Bank of America, Truist, U.S. Bank, PNC, and Navy Federal—representing a roughly 40% increase in mortgage accounts connected across the network since May. For customers using the Liabilities product, fill rates on key fields (interest rate, monthly payment, next due date) are now 97–99% across these servicers.
  

• 💳 Power richer credit card insights with APR data: APR data is now available for Chase, Citi, Wells Fargo, American Express, and U.S. Bank. Credit card data coverage continues to deliver near 100% fill rates across all key fields—balance, limits, statement balances, and payment amounts and dates—across all major issuers. 
  

• 🔍 New broker support in Investments: Plaid has added support for Stifel and Public, expanding the breadth of investment accounts your users can connect.
  

• 📊 Power smarter tax and portfolio insights with tax lot data in Investments: Plaid now returns tax lot information through the Investments API, starting with E-Trade, Interactive Brokers, and Robinhood. Tax lot data — including position type, cost basis, and quantity per lot — gives customers the granular detail needed to power tax-loss harvesting, portfolio optimization, and more precise net worth calculations.

Featured products: Sequential Foundation Model, Plaid Dashboard

What’s better:

• 🧠 Introducing Plaid’s sequential foundation model: Plaid's new sequential foundation model learns how financial events combine, in what order, and over what cadence to reveal how behavior unfolds over time. Where a flat summary might show two users with identical income, balances, and spending, the sequence tells a richer story. In early testing, the model prevented 26.5% more dollar value in payment returns at a fixed intervention rate, and reduced default risk by 13.6% at a 70% approval rate. 
  

• 📊 Troubleshoot institution issues with real-time visibility into your own traffic: Get real-time visibility into institution health for your specific customer traffic directly in your Plaid Dashboard. Institution Insights shows customer-specific success rates and error breakdowns filtered to your own traffic (not global metrics), drillable issue details with root cause and resolution guidance, and the ability to follow and receive updates on specific issues. You can use this feature for more personalized and actionable troubleshooting and observability.

• 🛠️ Troubleshoot and resolve issues faster in the Plaid Dashboard: We've revamped the support flow to make it easier to troubleshoot and resolve issues without leaving the Dashboard. The updated experience includes a streamlined UI with fewer page transitions, richer Known Issues content with root cause details and resolution guidance, and Bill—Plaid's chat interface—now available as an optional resource throughout the support process. You can also filter between Known Issues you're following and open support cases, and narrow by financial institution. 

New account fraud was the only major fraud typology to grow in both victim count (+31%) and dollar losses (+13%) last year. Fraudsters increasingly use generative AI tools to alter ID documents in ways that are hard to spot by eye, or by an analyst working a review queue. 

That's why we've shipped a number of Plaid Identity Verification (IDV) updates that strengthen every stage of the verification lifecycle: document verification, PII verification, and case management. Here's what's new and how it solves real-world problems.

Document verification: spot AI-manipulated images behind new account fraud 

Detect manipulated images of ID documents

IDV’s integrity check inspects the metadata to flag tampered images from virtual camera software and signs of digital editing or AI generation. This feature is designed to counteract fraudulent ID documents that are especially prevalent at the point of account opening. This feature flagged 50+ virtual camera attempts in one month alone.

But hey, fraudsters don’t always go high-tech. They can show paper copies or photos of stolen IDs instead during online account opening processes as well. In those scenarios, Plaid also detects screens or printed copies as another layer of protection against document fraud.

Require both sides of an ID document

Requiring both sides of the ID to be uploaded creates a more confident assessment. If only the front of a driver's license is uploaded, users are sent back to the document type selection step with clear feedback. This closes a loophole that fraudsters exploit and helps organizations who have stricter compliance rules, such as financial institutions, to reduce manual outreach to chase down what’s missing.

Data source verification: recognize trusted emails and rescue typos for genuine customers

Perform additional email risk checks

Plaid IDV uses a LLM that cross-checks a user's name and date of birth against their email address to identify mismatches. This doesn't fail users outright, but flags suspicious patterns as a high-quality signal. In its first two weeks, it caught 100% of fraudsters in a live attack on a major customer. Applicable to the dashboard and the API.

Detect typos and offer retries

When a user enters an incorrect DOB, SSN, or name accidentally during the account opening process, they'll have the option to correct the typo rather than failing the session outright. This improves pass rates to capture sessions that would have otherwise failed based on typos.

Case management: more intel and better labels in one place

These product updates are about giving investigators the full picture faster, without asking them to trade thoroughness for speed.

Visualize anomalies on a map

In the IDV dashboard, you’ll see a session's location signals plotted on a customer location map — Document Address, Submitted Address, IP Address, and IP Address (VPN) — showing the distance between them and an associated risk level. An IP address from an unexpected country, or a large gap between a submitted address and an actual connection point, becomes immediately visible instead of buried across separate fields.

Submit more granular fraud labels

When you mark a session as “fraud” in the IDV dashboard, you’ll see expanded fraud report labels. This helps with faster categorization, cleaner investigation trails, and better training data going forward. The expanded fraud labels include:

• User account and identity

  • User account takeover

  • Stolen identity

  • Synthetic identity

  • Multiple user accounts

  • Scam victim

• Bank account or payment method

  • Bank account takeover

  • Bank connection revoked

  • Card testing

• Transaction

  • Unauthorized transaction

  • Card chargeback

  • ACH return

  • Dispute

• Behavior

  • 1st party fraud

  • Missed payment

  • Loan stacking

  • Money Laundering

A configurable risk strategy for end-to-end coverage 

These Plaid IDV enhancements strengthen each stage of the fraud prevention journey – from onboarding to case management to risk configuration. The beauty is that these are self-serviceable, don’t require any additional integration work, and are live today. 

As you’re optimizing your risk strategy, ensure that you adopt a defense-in-depth strategy. Use a combination of data source verification, document verification, liveness checks, behavioral analytics, device fingerprinting, and network intelligence as trip wires to detect fraud, while fitting your risk appetite. 

Want to see these updates in the dashboard or discuss how they fit your risk strategy?

I spent nearly four years as FDIC Chairman watching open banking in the U.S. evolve in real time. Through painstaking work, financial institutions, data access platforms, and the fintechs consumers use to manage their financial lives online built trust across the ecosystem and confidence in it. That work produced something remarkable: consumers gained freedom to use their financial data with the products and services they choose. 

Consumer permission, combined with companies’ agreement to protect that data, is what has allowed digital finance to scale and why access remains open. Consumers gained the ability to securely connect their financial accounts to the apps and services they want to use, financial institutions gained confidence that consumers’ data would be handled within a framework designed for protection, and developers gained the ability to build entirely new categories of financial products. That combination of consumer choice and consumer protection is what made open banking successful. 

Now, AI is creating the next major shift in that story.

Consumers are increasingly connecting their financial data to AI assistants, building personal financial agents, using AI to better understand their spending, planning for major life decisions, and automating parts of their financial lives. They are actively seeking out these experiences because they create real value.

As financial data increasingly flows through AI and the tools evolve, the industry faces an important question: how do we ensure that consumer protections evolve alongside consumer choice? Consumers expect their financial information to be secure, and they expect control over where it goes. Those expectations become even more important when a new technology is involved.

That is why Plaid is actively working with customers, AI providers, financial institutions, and others across the ecosystem to ensure emerging AI-powered experiences can develop in ways that expand both consumer choice and protection. 

We believe consumers should be able to use their financial data with the AI tools they choose and trust. Today, Plaid makes it possible for consumers to share their financial data with hundreds of AI-native apps and services, including OpenAI and Perplexity. As customers of Plaid, all of these companies agreed to the same data protection standards that financial institutions have required of everyone in the ecosystem. 

Every app that helps consumers manage their money, every service that builds on open banking, every tool that uses consumer financial data to deliver value depends on trust in the ecosystem. That trust is built on strong security practices, safeguards, and industry standards that help protect consumers if something goes wrong. When consumer financial data flows outside of these frameworks, those consumer protections must travel with it. 

Through industry cooperation and collaboration, consumers gained access and control over their financial data with more freedom than ever before. We built that together with financial institutions of all sizes, with customers, and with advocates for consumer data rights. As AI reshapes the landscape, protecting that progress is work worth doing – and it is how we ensure the next generation of financial innovation is both more powerful and more trustworthy.

Finance looks like a ledger, but it behaves like a language. Transactions have semantics. Financial sequences have grammar. When you read the language, you can build intelligent financial products that actually understand people and adapt to their financial lives.

Plaid builds models to improve how financial services work, making financial products more personalized, useful and safe. Earlier this year, we introduced a transaction foundation model that understands the meaning of individual financial events, powering enhanced categorization, entity recognition, and income detection across many of our products.

Now, our sequential foundation model is learning the grammar of money: how financial events combine, in what order, and over what cadence to reveal how financial behavior unfolds over time. Plaid's network spans thousands of financial institutions and millions of consumers, giving the model exposure to a wide variety of financial formats, income levels, and economic behaviors, encompassing the breadth needed to learn patterns that generalize across real financial lives. For fintechs and enterprises building on Plaid, understanding these patterns and reading the language of money is what separates a confident lending decision from a costly one, or a fraudulent transfer caught early from a missed one.

Why sequence matters

Consider two consumers with nearly identical 90 day summaries: same monthly income, same average balance, same rent amount, same number of overdraft fees, and similar category-level spending.

For the first consumer, income arrives, rent and utilities clear, discretionary spending follows, and a single overdraft appears after an unusual home repair expense. The account recovers on the next paycheck and returns to its prior rhythm.

For the second consumer, the same events appear in a different order. A paycheck lands, a short-term loan repayment and credit card payment drain the account within 24 hours, small transfers bridge the gap, an ACH debit hits before the next payroll deposit, an overdraft fee posts, and the cycle repeats two weeks later.

A flat feature set may see similar income, balances, spend, fees, and repayment activity. The sequence tells a different story: one user experienced a temporary shock and recovered; the other is caught in a recurring liquidity squeeze.

For almost every use case that depends on understanding financial behavior, the most valuable signal is in transactions' order, cadence, and relationship to one another:

• For credit risk and underwriting, the overall composition of a borrower's financial life often matters more than their current balance. Better behavioral signals mean lenders can make more informed decisions, approving creditworthy borrowers while reducing defaults. 

• For ACH payment risk, the signal is rarely a single blatantly bad transaction, and more likely a transaction that looks wrong compared to past observed patterns. Better risk signals here mean catching fraud before it even happens.
  

How our sequential foundation model works

The model represents individual financial events using several complementary sources of information. One captures the meaning of the transaction itself through representations learned from large-scale financial data. Another captures timing, including the order of events, the spacing between them, and recurring calendar patterns. A third incorporates available account and transaction attributes such as amounts, account characteristics, and other contextual signals. These inputs are combined into a unified representation of financial activity over time. Rather than viewing a transaction as an isolated event, the model learns to interpret it within the broader flow of a user's financial behavior.

For example, a $100 transfer may reflect routine saving behavior when it follows payroll deposits and regular household spending. The same transfer can carry a different meaning of potential financial strain when it appears alongside overdraft activity or short-term liquidity events. Context allows the model to distinguish between superficially similar transactions that arise from very different underlying situations.

These event representations are processed by a sequence foundation model designed to learn relationships across a user's financial history. Instead of relying on manually labeled examples, the model is pretrained using self-supervised learning tasks derived directly from transaction sequences.

Several complementary objectives are trained jointly to help shape the representation:

• Contrastive Predictive Coding (CPC) encourages the model to learn patterns that make future financial activity more predictable, helping it capture recurring routines, income cadence, and evolving behavioral trends. 

• Replaced Token Detection (RTD) trains the model to recognize events that appear plausible on their own but are inconsistent with the surrounding sequence, improving its ability to identify anomalous or potentially risky activity. 

• Temporal Contrastive Learning (TCL) reinforces stable behavioral characteristics by encouraging representations from different periods of the same user's history to remain more closely aligned than those from different users.

Taken together, these objectives teach the model distinct but reinforcing skills: anticipating how financial behavior is likely to evolve, recognizing activity that appears out of place in its broader context, and maintaining a coherent understanding of a user's long-term financial patterns. The resulting representations capture both short-term dynamics and persistent behavioral signals, providing a foundation for downstream applications such as risk assessment and cash flow intelligence.

What this unlocks: early results

The sequential foundation model is already in testing with two of our most consequential use cases, and early results suggest it captures behavioral signals that traditional models miss.

For ACH payment risk, the sequential model prevented 26.5% more dollar value in returns at a fixed 1% action rate. For our customers, that means catching more risk without flagging more legitimate transactions, creating a better experience for end users and less revenue impact from returns.

For credit underwriting, we reduced the default risk by 13.6% at a 70% approval rate. Same volume of loans, meaningfully less risk.

These gains come without designing new bespoke product features. As Plaid incorporates the sequential model into our existing products, the lift in risk detection and underwriting performance flows through automatically. Better outcomes for our customers and their users from the same integration.

From bespoke features to shared sequential infrastructure

Historically, building a behavioral risk or underwriting model has meant hand-designing thousands of summary features: rolling averages, days since last cash advance, fee-to-deposit ratios. These features collapse the sequence into a few numbers, losing order and timing, and are often repeated for every new use case with small variations.

A sequential foundation model changes that equation. Teams pretrain once on a large, unlabeled corpus of financial activity, then adapt that shared model to each task with a small amount of labeled data. Improvements to the foundation model propagate to every product built on top of it—so the behavioral intelligence available to customers gets better over time, compounding in the same way we already see with the transaction foundation model.

What's next

Today, the model understands the grammar of money to tell a financial story. The next chapter expands its vocabulary to tell the story more fluently.

Balance events show not just what someone spent, but what liquidity remained. Account connection events capture changes in financial footprint, since linking a BNPL app or short-term lender is itself a meaningful signal. Identity and product usage events become more informative when interpreted over time rather than as isolated facts. Each new event type increases the signal available to every downstream product, without customers having to integrate those events directly.

Together, the transaction and sequential foundation models help us understand people’s financial lives and make better decisions for them, enabling true intelligent finance. 

If you're building AI-powered financial experiences and want to go deeper on how these models power use cases like cash flow underwriting or ACH risk, we'd love to connect.

What UK and US consumers really think about AI and money

In April, we released our State of intelligent finance report for the United States, exploring how AI is reshaping the way Americans understand and interact with their finances. Now, we’re releasing the UK findings, which reveal that the same shift is happening across the pond: AI in financial services is moving from experimentation to expectation—and pushing the industry forward from open finance to intelligent finance.

Powered by open financial data and accelerated by AI, intelligent finance represents a shift from the passive display of people’s finances to active guidance with AI, helping them better understand their options, make more informed decisions, and take action with greater confidence. 

We partnered with The Harris Poll to survey thousands of UK adults to understand how this shift is playing out in their daily lives, asking them questions about AI adoption, trust, and expectations in financial services. The results reveal a shared baseline, as well as notable differences in how consumers view the role of AI in their personal finances.

A shared baseline, with meaningful divergence

The headline finding is striking: AI adoption for financial tasks is nearly identical across markets. 54% of UK consumers and 55% of U.S. consumers have used AI for financial tasks in the past 12 months. In near-identical percentages, both populations also want transparency when AI is involved in financial decisions, the ability to review important AI-driven choices, and strong accountability when things go wrong.

But beneath this shared foundation, the two markets diverge in nuanced ways that matter for anyone building financial products.

Americans are more bullish on growth; Brits are more grounded in daily utility

U.S. consumers show stronger momentum intent: 41% expect to increase their AI usage for financial tasks in the next year, compared to only 33% of Brits. Americans are also more optimistic about AI's potential to deliver fairer outcomes in decisions like loan approvals (37% vs 30%) and they’re more excited by the prospect of fintech apps using AI (46% vs 43%).

UK consumers appear to be adopting AI less as a sweeping transformation story and more as a practical layer for everyday money management: helping them budget, compare products, reduce stress, and understand what to do next. That isn’t skepticism—adoption is nearly identical to the U.S.—but it does suggest a slightly different relationship with the technology, one rooted more in day-to-day utility than near-term enthusiasm.

UK Gen Z and Millennials are driving expectations forward

Here's where things get interesting. While overall adoption is similar, younger consumers in the UK are outpacing their U.S. counterparts in both usage and intensity of expectation.

77% of UK Gen Z and Millennials have used AI for finance in the past year, compared to 72% in the U.S. And their expectations are sharper: 61% of UK younger consumers believe managing money without AI will soon feel “outdated,” versus only 54% in the U.S. They're also more likely to expect AI to save them time, reduce financial stress, and help them save and invest more.

This UK cohort is treating AI as an essential layer of modern financial life and will gravitate to products that deliver it.

Trust is built differently in each market

Both countries want guardrails, but differ in how they want to be reassured.

In the U.S., the top factors that increase comfort with AI-enabled financial apps are protection guarantees, reimbursement policies, and full transparency on AI use. In the UK, the leading reassurance is human approval for major financial actions.

This is a meaningful distinction: while U.S. consumers are looking for safety nets, UK consumers are looking for a hand on the wheel. For product teams, this suggests that supervised autonomy—AI that can act, but with clear human checkpoints—will resonate with particular strength in the UK market.

Context shapes expectation

Finally, U.S. consumers report having experienced financial friction, including credit denial and fraud, at a higher rate over the past year. This may help explain why U.S. consumers are more willing to adopt AI for efficiency, protection, and decision support, having felt the pain points more acutely.

UK consumers, operating in a somewhat less friction-heavy environment, may be adopting AI more as an enhancement than a remedy.

What this means for fintechs and financial institutions

The data points to a clear set of priorities for financial services providers operating in either market: 

• Move beyond analytics to personalized guidance. Displaying balances and transactions is the baseline. Consumers want tools that interpret that data and surface actionable insights, be they a spending alert, a better rate, or a nudge toward a goal. The expectation gap between what consumers want and what most apps deliver is still wide.

• Be explicit about how AI is working for your customers. Trust isn't built through privacy policies, but through products that behave predictably, explain their reasoning, and make it easy to override or review decisions. In the UK especially, human oversight for major actions isn't optional, it’s necessary

• Meet the new financial consumer before someone else does. Younger consumers in both markets are treating AI as essential. If your app isn't answering the questions they're asking, they'll get their answers elsewhere. Competitive advantage isn't about features; it’s about embedding intelligence into the core experience.

• Keep humans in the loop—your customers expect it. Even the most AI-forward consumers want supervised autonomy, not blind delegation. The willingness to let AI act is conditional on the ability to confirm, undo, and understand. Products that respect this will earn deeper trust and longer retention.

The bottom line

For financial services providers, the opportunity is the same in both markets: build products that turn AI from a feature into a foundation—one that delivers real value, earns trust through transparency, and keeps humans in control of the decisions that matter most.

Plaid has been helping build that foundation for more than a decade. From powering real-time financial insights to developing our first transaction foundation model for finance, we provide the global infrastructure that makes intelligent financial experiences possible at scale.

If you're thinking about what intelligent finance should look like for your customers, we're here to help.

Anahita Walia
Product Marketing Manager

This month, we shipped updates that help businesses verify accounts more flexibly, make more confident credit decisions, reduce identity verification friction, and improve open finance reliability. Highlights include API-based bank account verification for backend-only flows, expanded income and asset verification for home lending, a revamped Credit Dashboard experience, identity verification improvements for workflow setup and typo recovery, and more precise disconnect timing for open finance Items.

Featured products: Database Auth API

What’s better or new:

• ⚡ Database Auth, ⚡ Backend-only account verification: Verify manually entered account and routing numbers via API—no frontend changes required, with implementation possible in under a day.

• 🔐 Broader verification coverage: Leverages Plaid network data to verify bank account details across banks, credit unions, and neobanks, including long-tail institutions.

• 🏦 Fewer ACH returns: Supports offline and non-interactive payment flows with flexible account verification options that improve verification coverage.

Featured products: Home Lending Report, Plaid Income (CRA Income Insights), Credit Dashboard

What’s better or new:

• 🏠 Plaid Home Lending Report, now integrated with Encompass Consumer Connect (POS): Plaid’s Home Lending Report can now be embedded directly into the borrower application experience, bringing seamless digital income and asset verification into the systems mortgage lenders rely on every day. 

• 💸  Plaid Income (CRA Income Insights), now available in Home Lending Report: Lenders can now leverage cash flow-based income insights alongside verified asset data in a single PDF report, helping simplify borrower evaluation and support a more complete view of borrower financial health.

• 💻 Revamped Credit Dashboard UI: Our redesigned no-code Credit Dashboard now delivers a more intuitive experience for reviewing consumer-permissioned financial data, enabling lenders to leverage cash flow underwriting data without requiring technical integrations or engineering resources. Get started with the Credit Dashboard.
  
  

Featured products: Identity Verification

What’s better:

• 👯 Duplicate IDV Templates: Teams can now duplicate existing IDV Templates and Configure Workflows directly from the dashboard — a highly requested feature that reduces manual setup time when standing up new configurations.
  
  

  
  

• 🔄 Typo Retries: When a user enters an incorrect DOB, SSN, or name during KYC, they'll have the option to correct the typo rather than failing the session outright. This improves pass rates to capture sessions that would have otherwise failed based on typos.
  
  

• Know exactly when an Item will disconnect with PENDING_DISCONNECT The PENDING_DISCONNECT webhook already gives you a heads-up when an Item is scheduled to disconnect. Now it also tells you precisely when. A new disconnect_time field returns the exact date the disconnection will occur, so you can time re-engagement prompts more accurately and reduce unnecessary churn. This is a non-breaking, additive change with no integration updates required. See the docs for details.

Kathleen McGuirk
GTM AI Lead

Bruno Werneck
Partnerships Lead, Financial Management

When a consumer reaches out to a fintech support team, the question is almost always about something that's not working as expected. How can I link my bank? When will I receive my money transfer? My card was denied for this purchase, can I pay by bank? These are critical moments in the consumer journey, and today most of them route straight to a human agent who has to dig across dashboards and internal tools to piece together an answer.

Plaid and Fin, a leading AI agent for customer service, are bringing the trusted Plaid Link experience directly into the Fin AI Agent. Joint customers can now help their end users connect (or re-connect) their bank accounts securely, all within Fin Messenger chat. Customers can now submit their interest in joining the open beta. Fin on average drops first response times from ~38 minutes to ~2 minutes once reaching active usage. We expect the Plaid x Fin partnership to further reduce first response times and resolution turnarounds to seconds. This partnership marks another step in the shift from digital finance toward intelligent finance, where AI agents can help consumers get personalized support and take actions to improve their financial lives.

Powering smarter support for fintech

Plaid and Fin already share many joint customers, including fintechs, lenders, banks, and credit unions, where bank connectivity sits at the heart of the customer experience and support volume can spike if a user can't complete a connection. Support tickets can show up during onboarding, payouts, or purchases, moments where leaving the conversation to find a help article or wait for a human agent is the difference between a completed task and an abandoned one. 

Now Fin can launch Plaid Link natively in chat

The Fin AI Agent integration with Plaid addresses these issues and can reduce support tickets by surfacing the right action at the right moment, without the consumer ever leaving the conversation.

Trust is the foundation of financial services, and it starts with secure, reliable data. Plaid is the trusted standard for financial connectivity, and Fin brings trust to AI-powered support. Together, we're helping fintechs and financial institutions deliver instant and reliable customer experiences.

— Michael Carpenter, SVP of Sales at Fin

Open beta: features You'll See

The first release focuses on the most common bank-connection job in fintech: helping an end user securely link their bank account so they can complete what they came to do. A few scenarios this unlocks at launch:

• Bank account linking during onboarding. A new user applies for a personal loan and gets stuck trying to connect the bank account that will be used for payments and for transferring the loan funds. Fin recognizes the issue and surfaces the "Connect bank" journey within a chat prompt. The end user completes account linking via Plaid and continues the onboarding flow once Plaid returns a successful session.

• Self-serve account connection from any conversation. A credit union customer asks about setting up direct deposit for a payout they're expecting. Instead of being redirected to their dashboard or payment settings, Fin offers to kick off the bank linking flow right in the chat. The user connects their account and is ready to receive the payout, without ever leaving the conversation.

What's next

Bank linking is the first step. Plaid and Fin are collaborating on additional Plaid-powered journeys inside Fin, including restoring disconnected bank connections, completing identity verifications, and troubleshooting account funding or payout instructions, all without the consumer leaving the conversation.

For customers interested in joining the open beta for Plaid in Fin, please register your interest here and reach out to your Fin or Plaid contact. If you're building on Plaid and want to explore how AI can improve your consumer experience, learn more here.

Digital identity is a cat and mouse game. And right now, the mouse is winning.

Three forces are reshaping how gaming operators handle identity. Players moved aggressively to multi-homing. Prediction markets opened a new front on the same player base. AI evolution made cross-operator fraud cheaper to scale. Most stacks haven't caught up. Identity in gaming used to be a per-operator problem. It isn't anymore.

Players multi-home. Operator stacks still assume they don't.

The engaged US gaming user now holds accounts at multiple platforms. A sportsbook, an iGaming brand, a DFS app, a sweepstakes site, and increasingly a prediction market on top. They shop odds across platforms. They chase promos across categories. They switch in-session.

At each new operator, the same legitimate player goes through the onboarding flow, re-registers, re-links a bank, re-uploads documents. Each new sign-up is treated as a blank slate. Is this person a VIP comfortably suited for high limits and frequency, a known fraudster, or somewhere in between? The operator has no way to know. The same person, treated as a stranger every time they cross a platform boundary.

That friction has a cost. More than half of Americans (58%) have already abandoned a financial app because it couldn't sync with their bank. When onboarding breaks, legitimate players bail. 

This is a CAC and retention problem before it's a fraud problem. Operators pay full acquisition cost for players the category has already converted, and the cost compounds with every product launch and every new market. As US OSB matures and acquisition costs climb, that compounding loss is becoming the most expensive blind spot in the gaming operator's economics.

The player is the same. The infrastructure to recognize them isn't.

Prediction markets opened a new regulatory front on the same player base

Prediction markets added another identity seam. Kalshi, Polymarket, Robinhood, DraftKings Predictions, and major gaming brands are pulling players into products that sit beside traditional sportsbooks but operate under different regulatory regimes.

That split matters for identity. Prediction markets operate under CFTC federal oversight, while sportsbooks follow state-by-state OSB rules. Different rule books mean different KYC standards, AML thresholds, verification flows, and reporting obligations. Same players. Same fraud actors. More seams to exploit. 

Players and fraud actors now move across regimes. The defense needs to move with them, with the behavioral, device, and IP signals that catch fraud and follow the actor.

AI made cross-operator fraud cheap to scale

Organized fraud rings have always coordinated across operators, and the unit economics used to make it expensive. AI changed the math. The cost to commit fraud dropped sharply. The procurement and compliance cycles that constrain defenders haven't.

The stakes are large. Global fraud losses are projected to reach $40 billion within the next few years, driven in large part by AI-enabled attacks.

In gaming, the cost asymmetry plays out across the network. Fraudsters cycle through identity and device combinations at multiple operators simultaneously, evading detection. AI-generated documents look more convincing than ever and can bypass simple KYC checks. Bot-coordinated account behavior runs across platforms in parallel. Synthetic identity rings that took weeks to spin up in 2022 are provisioned in hours.

Per-operator stacks see one slice and file it under background loss. A network-level view sees the same signature, the same timing, multiple platforms. The signal is loud if anyone is looking at the right altitude.

What a modern gaming identity stack should include

Most identity stacks were built in layers: KYC, device intelligence, behavioral analytics, transaction monitoring. Each layer is useful. But none is enough on its own, especially when fraud moves across operators. The missing anchor is the bank account: a persistent, permissioned signal tied to real financial history.

A modern gaming identity stack needs four things working together.

1. Adaptive risk decisioning across the network. No single operator stack sees the full picture of fraud. A risk engine grounded in cross-network behavioral and device patterns catches what individual stacks miss.

2. Bank-account ownership and history as the cross-operator anchor. Documents can be deepfaked. Devices can be spoofed. A real, funded, consumer-permissioned bank account tied to longitudinal financial history with real transaction patterns is structurally hard to fake at scale. It is the foundational authenticity signal that survives where document and device signals fail.

3. Risk signal at every moment, not just at sign-up. Registration, deposit, and re-verification, each moment requires its own decision. The behavioral and bank account signals captured at registration continue to inform decisions long after.

4. Defense-in-depth across the verification stack. No single check stops AI-driven fraud. Defenses need to layer at every stage of identity verification: capture, detection, verification, and decisioning. When fraudsters cycle through one stage to bypass it, the next stage catches them. A layered defense can entail virtual camera detection for manipulated images and facial duplicate detection or DMV checks to combat synthetic identity fraud. 

Where Plaid fits

Plaid helps gaming operators make better identity and risk decisions across the player journey: onboarding, deposit, re-verification, and fraud detection.

Plaid's network spans the financial ecosystem — more than 9,000 apps and clients across 12,000+ financial institutions, with 1 in 2 American adults using Plaid. That scale gives operators broader signals across bank accounts, devices, and account behavior, without requiring operators to share data with one another.

Across the player journey, that looks like:

• Player onboarding: One-click identity information onboarding and bank authentication at registration for players already in the Plaid network. Identity verification step-ups (document verification, liveness) trigger when higher risk is detected.

• Deposit risk: Real-time transaction risk scoring at funding, with optional ACH return guarantee for risk-immature programs.

• Returning player verification: Confirming the operator’s returning players (including dormant accounts and ATO defense) against linked bank-account ownership information.

• Network fraud detection: Cross-platform fraud signatures the individual operator can't see in its own data.

Plaid's solutions span regulated gaming — sportsbooks, DFS, online poker, horse racing — and prediction markets. The same fraud infrastructure operates at each one.

What good looks like in 2026

The operators winning in 2026 will be the ones that stop treating identity as a one-time compliance check and start treating it as infrastructure across the player journey.

That means matching friction to risk at onboarding, reducing unnecessary friction at deposit, and detecting fraud patterns individual operators cannot see on their own. The same network that helps legitimate players onboard faster also catches fraud rings earlier. 

Network-grounded signals reduce fraud loss. For gaming operators, those are no longer separate problems. They are two sides of the same infrastructure decision.

The operators that modernize now will be better positioned for the next wave of product expansion, regulatory scrutiny, and AI-enabled fraud.

Learn how Plaid helps gaming operators win with faster, more secure identity and fraud prevention.

Across industries, Plaid’s customers are growing faster than ever before. The average time for a Plaid customer to hit a million users onboarded has dropped by 33% which corresponds to nearly a full year faster. This applies to our customers across fintech, prediction markets, and agentic finance.

With every company growing faster and faster, you have to keep up. More and more money is being spent on marketing, product development, and sales. Unfortunately, in the backend, compliance teams are being overrun by hundreds of new users every day and each one needs a manual review where the cost of failure is in the millions of dollars.

An increasing tax on company growth

As we worked with our customers, we noticed a realization: our fastest growing customers all had the largest backlogs. The faster you grow, the more resources your team has to spend on training new compliance hires on tedious watchlist checks. 

As you post your launch video of your brand new product, your company blows up on X and customers sign up in droves, but your compliance team gets a 10,000 user increase in their backlog in a single day. These 10,000 users won’t get to access your product for a month. Not to mention they’re already at a higher risk of churn now and can be vocal critics of your platform. 

Each 1% in growth rate wasn’t just an increase in users but in fact a rapid increase in both AML backlog and another compliance team member that you need to onboard.

Overcoming the 30 day period

This backlog can’t just be solved in a single day, instead your compliance team struggles with the operations of managing a review queue that never really empties. The average time for a good user (false positive watchlist case) to be in a customer’s watchlist backlog is nearly 30 days. 

Beyond just the number of false positives, it’s difficult to handle the manual triage, context switching, and poor auditability. All of these issues are then compounded with the false positives from the “John Smith’s” around the world. We’re in an environment where companies are growing faster than ever, with lower head counts. Even worse, due to increasing global tensions, the number of individuals and entities added to sanctions lists has increased 30% over the past two years.  

This is the problem we've solved with our new Monitor redesign. We didn’t change what Monitor does – the matching algorithm, the watchlist coverage, and the automated rescanning are all still there and still best in class. Instead, we’ve reimagined an AML compliance program for the world where billion dollar companies can be built by just 10 employees. 

What’s new in Monitor 

How we safely speed up a compliance process 

We started from the ground up – counting every click in the flow. Making a decision on a watchlist hit required multiple clicks to gather context on lists flagged, what the match level was, what the current case status showed. We then had a single goal: decrease the number of clicks to as close as zero while still keeping a world class AML experience. 

The new dashboard brings all of it into a single view:

• The left panel displays the full pending queue, where analysts can filter by what's assigned to them, what is unassigned, or everything in the queue. 

• The top card surfaces the individual or entity and the review status

• The cards below show the current hit count and relevant watchlist matches at a glance. 

• A persistent decision review widget stays visible throughout, so analysts can leave a comment and submit without losing their place. 

This creates a higher analyst productivity and a better case management experience. For a team working through high volumes daily, fewer clicks per case makes a big impact — and clearer confirmation means fewer duplicate reviews eating into capacity.

When it comes to risk configurations, Plaid Monitor still allows you to configure program templates based on your risk tolerance levels. All program templates are still in one spot, but are now clearly organized by individual programs and entity programs.

What Monitor still does well

The dashboard design is new and optimized for efficient case management. But the foundational value of Monitor remains intact:

One tool, every major sanctions list covered 

Monitor screens customers against one of the broadest sets of global watchlists available to achieve AML compliance — OFAC, PEP lists, EU, UK, Australian, Canadian, and UN consolidated sanctions, Interpol Red Notices, and more — all updated daily.

A single view of identity and risk 

Unlike standalone sanctions screening tools that require a separate identity verification vendor, Monitor integrates natively with Plaid Identity Verification — giving your team a single, unified view of both identity and watchlist status. This not only addresses KYC compliance and customer due diligence needs, but creates a faster ramp for new analysts: one mental model for configuring KYC, AML, and fraud workflows.

"With case management, ongoing monitoring, and list provisioning, Plaid Monitor fits all of our needs and saves us from having to go to multiple providers."

Accurate signal, less noise 

The matching algorithm handles phonetic variations, transliteration across 21 languages, nicknames and aliases, and semantically similar names across scripts — keeping analysts focused on higher-risk alerts rather than chasing noise.

Always-on monitoring, audit trail included 

Daily automated rescans that cover your entire customer base without manual rescreening workflows – scaling to over 100 million users without performance degradation. When a new hit is found, your team is alerted via webhook in real time. Every decision is captured in a persistent audit trail — including workflows, user roles, and secondary review configurations — so you’re ready for regulatory examination.

Get started

AML compliance operations can get buried under operational friction that compounds as teams and customer bases grow. The updated Monitor dashboard is built to remove that friction — so the analysts you have can do more, and the decisions they make are better supported.

If you're already using Monitor, the new dashboard is live now. Log in to see the updated review experience and explore the new case management settings.

If you're not using Monitor yet, we'd like to show you what it looks like in practice. Talk to our team or get started today.

In 2025, the UK made 351 million open banking payments: a 57% jump on the year before, with VRP accounting for roughly 16% of all open banking transactions and growing 98% year on year. It’s clear that pay by bank works for recurring payments. 

Until recently, most of that growth was limited to sweeping, money moving between accounts a person already owns. Useful, but a fraction of what Variable Recurring Payments (VRP) can actually unlock. Now, the UK is opening up the real recurring payments market: from consumers to businesses, all regulated and at scale.

What VRP use cases are available, and why "commercial" is the unlock

Sweeping VRPs, or "me-to-me" payments have made a major impact. They’ve unlocked smart savings, overdraft avoidance, and instant fintech funding. 

With Commercial VRP (cVRP), UK businesses can use VRP to make it easier to transact with their customers. Wave 1 covers five high-trust sectors:

• Financial services. Payments into regulated products like savings accounts, ISAs, mortgages, pensions, and insurance, as well as FCA-authorised e-money accounts.

• Utilities and telecoms. Electricity, gas, water, broadband, and mobile.

• Rail. Tickets purchased under National Rail Conditions of Travel.

• Charities. Donations to organisations registered with the Charity Commission, OSCR, or CCNI.

• Government. Central and local government bodies, including TfL-designated organisations.

For an account funding use case like an investments brokerage this now means end users can: 

• Create a single consent with clear limits

• Securely verify it with their bank

• Make one-time or scheduled monthly account top ups under a single VRP consent

• Funds landing with the brokerage in near real-time

Wave 1: where safe recurring payments grow next

Wave 1 focuses on regulated, high-trust sectors where recurring payments are already common: regulated utilities, regulated financial services, central and local government, charities, and rail. These are sectors where consumer protection is already strong, transaction predictability is high, and the value of flexible recurring payments is most obvious.

cVRP introduces three things that direct debit and continuous card authorities do not:

Irrevocable, instant settlement. Once a payment executes, the funds move in real-time. There is no chargeback window, no indemnity claim. There is now a clear dispute framework. For clients, this means eliminating an entire category of financial and operational risk.

Granular, consumer-controlled consent. The consumer authorises the mandate with specific limits. You get instant confirmation once a mandate is set up and end users can view and revoke mandates at any time from their bank app. 

SCA at consent, not at payment. With VRP, strong authentication happens once when the mandate is created. Subsequent payments execute without friction, on the strength of that authenticated consent. This gives merchants the conversion benefits of stored credentials combined with the benefits of strong customer authentication. 

The security case is already proven. In the first half of 2025, open banking payment fraud was roughly 3.5 times lower than other payment fraud types. That’s largely due to how open banking payments are designed. They don’t share card numbers with the merchant, and every payment is authenticated directly through the payer’s bank. VRP extends this to recurring payments, replacing stored card credentials with a single bank-authenticated mandate.

The trajectory ahead

cVRP opens the addressable market by an order of magnitude: every consumer with a recurring relationship to a regulated utility, financial services provider, or public sector body in the UK is now in scope, with more use cases going into effect later this year. And, as the nature of transactions changes and agentic commerce takes shape, VRP provides the regulated, consumer-controlled rails to support it.

How to get started using VRP with Plaid

Plaid has been live with VRP since 2022, and as a founding participant in the UKPI scheme, we are ready for commercial VRP from day one. 

VRP is built on the same Payments API that powers Plaid's one-time Pay by Bank, so if you're already integrated, adding recurring payments is straightforward. And because VRP sits alongside our identity, account data, and risk products, you get a single provider for the full stack — from verifying a customer's identity and affordability through to initiating and reconciling payments. With over 9,000 apps on the Plaid network globally, each transaction also benefits from network-level fraud signals drawn from user behaviour across the ecosystem.

Lenders across the UK and Europe are facing a growing challenge: traditional credit data alone no longer captures the full picture of a borrower’s financial health. 

As variable income, gig work, and multiple income streams become more common, many borrowers no longer fit neatly into traditional underwriting models. For lenders, this creates a difficult trade-off between approving credit invisible borrowers or turning away potentially qualified applicants. The impact is already being felt across the market, with nearly 70%* of UK gig workers reportedly struggling to access financial products. 

At the same time, responsible lending requirements are becoming more stringent, with frameworks like the FCA’s Consumer Duty and Consumer Credit Directive (CCD2) raising the bar for how lenders assess and document borrower affordability. To navigate today’s evolving credit landscape, lenders need more comprehensive and real-time insights into borrower income and affordability. 

That’s why we’re excited to announce Plaid Income in the UK and Netherlands, our solution for real-time income verification powered by consumer-permissioned financial data, with expansion across Europe planned for later this year.

A better way to verify income

Traditional income verification methods, like payslips, manual uploads, or bureau data, can introduce friction in the application process and often fail to reflect a borrower’s current financial reality – especially for consumers with variable or non-traditional income.

Bank-derived income verification offers a more real-time view of affordability by using consumer-permissioned transaction data from connected bank accounts. Instead of relying on static snapshots of financial health, lenders can better understand how borrowers earn, spend, and manage cash flow over time.

Plaid Income takes raw deposit activity and transforms it into structured income signals lenders can use to make more informed affordability assessments.

It starts with a simple account connection

Borrowers connect their bank accounts and grant permission through Plaid Link, Plaid’s secure, consumer-facing account linking experience. By making account linking fast and intuitive for borrowers, Plaid Link drives up to 90%** conversion within lending flows, meaning more borrowers successfully complete the verification step.

Structured cash flow insights for confident lending decisions

Once connected, lenders gain access to up to 24 months of categorised bank account data – inflows, outflows, balance trends, and account ownership details, including a Primary Account Indicator that helps identify the account a borrower uses most actively. Rather than relying on a simple ownership check, the indicator uses signals like transaction frequency and volume, account balances, recurring income and expenses, and recency of activity to determine which accounts are most relevant for affordability assessment.

This gives lenders greater confidence that they are evaluating the accounts most representative of a borrower’s day-to-day financial activity, while also supporting the documented affordability requirements increasingly expected by regulators.

Smart income categorisation and forward-looking intelligence

Raw transaction data alone can be difficult to interpret, particularly for borrowers with variable or non-traditional income.

Plaid Income classifies earnings across income types like salary, gig work, self-employment income, and benefits, helping lenders more accurately assess borrowers whose income may fluctuate month to month. For salaried workers, Plaid estimates income using recurring net deposits. For borrowers with variable income, Plaid applies forward-looking forecasting models designed to account for changing income patterns over time.

Plaid Income also surfaces additional signals like pay frequency and predicted next payment date, giving lenders a more structured view of how and when borrowers get paid.

Open banking connectivity through a single integration

Plaid Income is built on Plaid’s global open banking infrastructure, helping lenders securely access consumer-permissioned financial data across thousands of financial institutions through a single API integration.

This broad coverage helps lenders scale income verification across markets without managing multiple providers or integrations. Once a borrower connects an account, lenders can retrieve income insights and refreshed reports directly through the API, without requiring borrowers to reconnect their accounts.

For lenders already using Plaid in the US, expanding into the UK is a natural extension. For UK and Netherlands-based lenders, the same integration will work across Europe as we expand later this year.

Get started with Plaid Income for Europe

As income patterns become more dynamic and responsible lending expectations continue to evolve in our region, lenders need better ways to understand a borrower’s financial reality.

By transforming consumer-permissioned bank data into real-time income insights, Plaid Income delivers a modern income verification solution that supports faster, more informed lending decisions.

Plaid Income is now available for early access in the UK and the Netherlands, with broader availability across Europe planned later this year.  If you're a lender looking to get started, or a partner looking to build with us, contact your Plaid account team to join the waitlist.

*Credit Connect, 2023; The Fintech Times, 2023

**Conversion rates vary by lender and implementation. Figures based on top-performing integrations across Plaid Europe’s lending customer base.

Zach Perret
CEO & Co-Founder, Plaid

We just wrapped up Plaid Effects, our annual customer conference. It was great spending time with customers and hearing what teams are building. One theme came up in nearly every conversation: AI is rapidly reshaping financial services.

We believe the next generation of financial products will require a financial intelligence layer that can understand financial activity at scale. Today at Effects, we announced a new set of analytics products, and foundation models designed to help power that future: across fraud, credit, payments, and financial management. Here’s what we shipped.

New foundation models

We announced a set of foundation models trained specifically on large-scale financial data from the Plaid Network, and using them as the common intelligence layer across our products.

First, we announced a foundation model that provides a deeper understanding of transactions data across Plaid - not just what happened, but why. With this model, we’re seeing a 13% improvement in categorization accuracy while income classification is up by 48%.

Now we’re working on a new model that understands financial activity as a continuous narrative. Instead of looking at transactions as isolated events, this model looks at the sequence of transactions to determine what financial behavior means in context, and how it's changing.

Collectively, these models are an intelligence layer shared across our products so improvements means your products are getting better without you having to do anything else.

Fraud detection beyond your four walls

We announced Trust Index 3 (Ti3), the latest machine learning model powering Plaid Protect. Ti3 uses a dramatically more sophisticated fraud graph to identify modern fraud patterns: including synthetic identities, account takeovers, and coordinated fraud rings, by tracing relationships across devices, identities, bank accounts, and applications in real time. In testing, Ti3 catches up to 41% more fraud at the same false positive rate.

Lending built on real-time financial behavior

We rebuilt Plaid Income with a proprietary transformer-based LLM that develops a deeper understanding of the context behind transactions, identifying earned income with far greater accuracy, achieving 86% precision without sacrificing recall.

We also added new cash flow and network attributes to LendScore, Plaid’s credit risk score, to improve underwriting performance, analytics depth, and adverse action explainability. LendScore improves predictive performance by 25% compared to traditional credit data alone and can reduce risk by up to 41% while maintaining approval rate.

Lastly, lenders can now subscribe to refreshed borrower data over time using the same endpoint they already use for underwriting and income. This gives them insights into borrower financial health after a loan has been originated so they can reassess risk and proactively manage changes over time.

Smarter decisioning for earned wage access

We also introduced the Plaid Cash Advance Index, a new risk model purpose-built for cash advance and earned wage access providers. These products involve high-frequency, real-time underwriting decisions, but most providers still rely on signals from a single connected account. Cash Advance Index uses network-level intelligence to predict likelihood of repayment within 30 days, helping providers approve more of the right users, extend more appropriate advance amounts, and manage risk more precisely.

In a randomized A/B test with a leading provider, the model reduced delinquency by eight percentage points with no reduction in approval rates.

Instant bank payments without risk

We launched Plaid Guaranteed Payments to help companies approve more transactions while managing risk. With Guaranteed Payments, Plaid evaluates every transaction with our risk model, if approved, we guarantee settlement. If a payment fails, we cover the loss and handle recovery.

Companies using Guaranteed Payments are already seeing approval rates as high as 90% on instant funding flows, with implementations in as little as two weeks.

Developer tools built for the agentic era

The way developers build has changed. A single developer with an AI coding agent can now ship what used to require an entire team, so we rebuilt our developer tools for that reality.

We launched Sandbox Studio, a new environment inside the Plaid Dashboard that lets developers quickly generate test users, run scenarios instantly, and inspect live responses in one place. We also introduced the Plaid CLI and a collection of MCP servers so AI agents can interact directly with our sandbox programmatically.

Later this year, we’ll launch an MCP server for banks and financial institutions as well, enabling data partners to manage Plaid integrations directly from tools like Claude and Cursor with automated validations, debugging, and AI-suggested fixes.

Let's build the future of intelligent finance together

Plaid started by building the connectivity layer powering thousands of financial applications and services. Now, we’re building the intelligence layer on top of that network - helping companies understand financial behavior in real time and build more personalized, secure, and intelligent financial products.

Thanks to every customer, partner, and builder who joined us this week and continues pushing the industry forward. Many of the products we announced today are available now or in early access. Reach out to your Plaid account team to learn more.

There’s a lot more coming. Check out the rest of the talks from Plaid Effects.

ACH is the backbone of the digital economy and one of the most cost-effective ways to move money, growing 17% over the last five years. But running an ACH program comes with unique risks that brands need to manage.

Settlement and return windows can stretch for days, putting risk managers in a bind: enable the experience immediately and take on the exposure, or wait until the risk is lower and make the user wait. As volume grows, so does the cost of making the wrong tradeoff.

Teams approach this challenge in two different ways.

Some have built ACH expertise in-house and want scores and attributes they can plug into their own decisioning, so they can control every threshold. For the past five years, Plaid Signal has been the answer: an AI-powered ACH risk model trained on $230B in transactions that surfaces account and usage insights from across the Plaid network. Insights that no individual customer can see from inside their own four walls.

Other teams just want answers. They don’t want to build a risk program right away. They simply want bank payments to work so they can focus on their core product.

Introducing Plaid Guaranteed Payments

Plaid Guaranteed Payments is a fully managed bank payments solution built for those teams. Plaid evaluates every transaction in real time and returns an approval decision in milliseconds. If Plaid approves it, we guarantee settlement. If a payment fails, we cover the loss and handle recovery.

That means you can fund your customer’s account instantly, credit repayments faster to allow further drawdowns, or deliver your product or service the moment a customer says yes. Your team never sees a return and your customers get a seamless payment experience.

Plaid Guaranteed Payments also adapts to your type of risk. Get full coverage across all eligible returns, or focus on the specific risk types that matter most to your business. You only pay for the protection you actually need.

Approval rates as high as 90%

Running a best-in-class risk program requires deep expertise, a dedicated team, and continuous model improvement. With Plaid Guaranteed Payments, our models and network intelligence work on your behalf. 

The teams using Plaid Guaranteed Payments are seeing approval rates as high as 90% on instant funding flows, with implementations in as little as two weeks. 

Decisions powered by the Plaid network

That level of performance is only possible because of the data feeding the decisions. Plaid Guaranteed Payments is built on Signal, our proprietary AI-powered ACH risk platform, and Protect, our real-time fraud intelligence system.

Plaid Signal has been trained on $230B transactions across millions of accounts from 12K+ financial institutions, and uses insights and network activity from an account to determine ACH return risk. Plaid Protect identifies patterns across 500M+ linked accounts and a billion devices to detect user fraud. Together, this gives Plaid a full picture of payment risk based on account activity and user behavior across the broader financial ecosystem.

Instant certainty for bank payments

Bank payments have always meant a tradeoff: low cost, but days of uncertainty. That's changing. With real-time decisioning and guaranteed settlement, teams have something they've never had with ACH before: instant certainty. Customers get their funds, services, or goods the moment they say yes. Teams stop iterating on risk models and get back to building.

Bank payments, finally working the way you'd want them to.

Get started

Plaid Guaranteed Payments is now available through Plaid Transfer. To learn more, contact your Plaid account team or join the waitlist.

Today, we’re announcing Trust Index 3 (Ti3), the latest version of the machine learning model powering Plaid Protect. Ti3 expands how Protect detects fraud across the Plaid network with a dramatically larger fraud graph, deeper real-time relationship analysis, and new signals designed for modern attack patterns.

The result is a fundamentally different approach to fraud detection, one built on visibility across the network, not just within users from a single platform. Early testing results are promising: Ti3 catches up to 41% more fraud while maintaining the same false positive rate as previous models.

A larger graph built for coordinated fraud

At the core of Ti3 is a major expansion of Plaid’s fraud graph, a network map of how devices, accounts, identities, sessions, and institutions connect across the financial ecosystem.

By understanding how those entities relate to one another over time, Ti3 can identify coordinated fraud patterns that would otherwise remain invisible inside a single app or transaction.

Our graph has grown nearly 10x in just a matter of months, dramatically increasing the number of relationships we can analyze across devices, accounts, institutions, sessions, and identities.

Ti3 can also now traverse that graph up to nine hops deep in real time, essentially following multiple degrees of separation between devices, accounts, identities, and sessions to uncover coordinated fraud activity that would otherwise appear unrelated.

That means Ti3 can identify patterns like:

• A device opening accounts across multiple platforms in rapid succession

• Shared infrastructure linking seemingly unrelated identities

• Repeated account linking and disconnection behavior across the network

• IP locations that are geographically inconsistent with the addresses tied to connected bank accounts

• Device and session activity that deviates from established account behavior patterns

Often, the difference between stopping one fraudulent event and uncovering an entire fraud ring comes down to visibility across the network.

New signals designed for emerging attack vectors

More actionable fraud attributes

Ti3 introduces new signals and attribute categories designed to address increasingly sophisticated fraud patterns, while making the underlying insights easier for fraud teams to understand and act on.

As attacks evolve, fraud teams need more than just accurate risk scores. They need clearer signals that explain why activity appears suspicious. With Ti3, we’ve expanded and refined the attributes powering Protect to surface more actionable intelligence across account behavior, device activity, identity patterns, and network relationships.

More accurate account age intelligence

Synthetic identity fraud and bust-out schemes often depend on making new accounts appear established. Ti3 combines bank account age, transaction history, and network connection patterns to better distinguish genuinely tenured accounts from recently created ones.

Enhanced device identification

Account takeover attacks have become increasingly sophisticated as fraudsters rotate devices and spoof fingerprints to evade detection. Ti3 introduces enhanced device identification powered directly through Plaid Link, enabling more precise identification that is significantly harder to bypass than traditional device fingerprinting approaches.

Built for the future of fraud detection

For over a decade, Plaid has served as infrastructure connecting thousands of financial applications and the banking system. Every linked bank account, authentication event, and device interaction contributes to a broader understanding of how fraud moves across the ecosystem. 

Nearly one million people connect through Plaid every day.

That network visibility enables fraud intelligence that is difficult to replicate because it was built over years of real-world financial activity, not assembled after the fact.

But the advantage isn’t just the scale of the network. It’s the speed at which we improve it.

We treat fraud prevention as a race, not a finish line. Ti3 is the latest result of a continuous model enhancement strategy where we ingest, analyze, and deploy ecosystem-wide intelligence at a velocity the rest of the industry struggles to match.

And we’re continuing to invest in what comes next.

To learn more about Protect and Ti3, reach out to our team today.