Fraud Detection & Transaction Risk
S

Stripe Radar

Stripe Radar is Stripe's machine learning fraud product for businesses that take payments online and for platforms and marketplaces with connected accounts. It scores every payment in real time with models trained on more than 70 trillion data points from the Stripe network, which processed $1.9 trillion in payments in 2025, and it runs by default on Stripe payments. An API also scores payments, accounts and customers for businesses that process elsewhere, so a team can keep its existing processor.

Coverage runs across the customer lifecycle: card testing and transaction fraud at payment, risky accounts and merchant monitoring at onboarding, and abuse such as multiple accounts, free trials, bots and account sharing. A business can allow or block a payment, trigger Adaptive 3D Secure or send it to manual review. Paid plans add risk scores from 0 to 99, custom rules backtested on past payments, risk preferences and Radar Assistant, and Pro adds dynamic risk thresholds. Dispute prevention draws on Verifi and Ethoca.

Radar Lite is included with standard Stripe card pricing. Radar Standard, Plus and Pro cost $0.05, $0.07 and $0.09 per screened transaction pay as you go, or start at $10, $14 and $20 a month with 200 transactions included. Platform plans start at $20, $44 and $70 a month, and enterprise access is quoted.

Stripe reports a 32% average reduction in fraud. FreshBooks blocked more than 300 fraudulent accounts from onboarding in three months, Anthropic saw 83% fewer legitimate transactions wrongly blocked, and DoorDash cut chargeback costs by 10% on payments processed outside Stripe.

Last VerifiedNovember 20, 2026
Compare Stripe Radar with other vendors
Founded
—
Headquarters
—
Categories
fraud-and-transaction-risk
Assessment

Capability Axes

Capability grades

15 of 15 axes rated · 14 graded A or B

AI Capability
AI Centrality
AA on AI CentralityThe artificial intelligence is the product. Remove the models and there is nothing left to sell.
Vendor Published

The learned risk signal is what Radar sells. Adaptive models trained on the Stripe network score each payment, account and customer from payment signals and outcomes such as fraud that merchants report, and the same signal is sold on its own to businesses that process elsewhere. Custom rules let operators layer their own policies on top.

Autonomy and Oversight Model
AA on Autonomy and Oversight ModelWhat the system runs alone, what constrains it, and how a person checks it are all published: modes, thresholds, sampling or audit controls, and the route a case takes to human review.
Vendor Published

Each payment can be allowed, blocked, sent to 3D Secure or routed to a review queue. Businesses on Plus or Pro write rules, backtest them against earlier payments and set risk preferences, and Pro adds dynamic risk thresholds. Reviewers inspect queued payments in the Dashboard where the payment method supports it, and for connected account decisions the platform must give the affected account a route to human review.

Model Risk Management and Transparency
BB on Model Risk Management and TransparencyReal transparency mechanisms are published, such as per alert explainability, confidence scoring or split testing, without the validation package or supervisory mapping behind them.
Vendor Published

Paid plans show a 0 to 99 risk score beside the highest, elevated or normal risk level, and custom rules can be backtested on past payments before they go live. Stripe evaluates model changes on precision, recall, authorization and block rates, in aggregate and per merchant, and monitors performance after release as fraud patterns shift. A version history for production models and an independent benchmark by payment method are not published.

Operational and Outcome Evidence
AA on Operational and Outcome EvidenceNamed customers with hard performance figures and enough method to test them.
Vendor Published

FreshBooks blocked more than 300 fraudulent accounts from onboarding onto its payments service in three months, combining merchant information with transaction monitoring. Anthropic saw an 83% reduction in legitimate transactions incorrectly blocked, and DoorDash cut chargeback costs by 10% on payments processed outside Stripe, by Stripe's account. Across its users, Stripe reports a 32% average reduction in fraud.

AI Safety and Data Stewardship
BB on AI Safety and Data StewardshipA categorical stewardship commitment is published without the retention schedule or the engineering detail behind it.
Vendor Published

Network learning is the design: Stripe pools transaction signals across its merchant network, so one business's activity helps score payments for others. The Privacy Center limits that use to preventing fraud and improving payment performance, not judging an individual's character or creditworthiness, and uses aggregated or deidentified information so a business cannot identify another business's customers. The Radar terms require merchants to delete certain Radar data at termination or on request. Retention periods for the training data are not stated.

Regulatory and Compliance
GLBA and Data Privacy Posture
BB on GLBA and Data Privacy PostureA substantive privacy document that reaches the product itself, short of the subprocessor list or the full data handling detail.
Vendor Published

A data processing agreement sets Stripe's processor duties, safeguards and transfer terms, and Stripe holds CBPR and PRP privacy certifications and participates in the EU-US, UK and Swiss data privacy frameworks. The Radar terms require businesses to tell their customers that Stripe collects device characteristics and customer activity indicators for fraud detection, and the Privacy Center explains how Stripe uses personal information in its own fraud models.

Security Certifications and Trust Center
AA on Security Certifications and Trust CenterCertifications named with their type and presented as retrievable artifacts, usually through a trust portal a buyer can open without asking.
Vendor Published

A SOC 3 report downloads freely from Stripe's security page. Coalfire Controls examined the Stripe Global Payments and Revenue and Finance Automation system for October 2024 to September 2025 against the security, availability and confidentiality criteria, and the system description names Radar in scope. SOC 1 and SOC 2 Type II reports are produced annually and come through the Dashboard or on request, and Stripe is certified as a PCI Service Provider Level 1.

Regulatory Status and Licensure
BB on Regulatory Status and LicensureThe regulatory position is clearly stated and appropriate to the product, with part of the verification left to the buyer.
Vendor Published

Stripe's terms present Radar as a risk service, leaving the business to apply its own decision rules and to handle the payments it accepts. Stripe Payments Company holds state money transmitter licenses for its money transmission, and those licenses sit with that entity and activity rather than with the fraud product. Radar for Platforms extends the same risk information to platforms deciding on their connected accounts.

AI Governance and Bias Disclosure
BB on AI Governance and Bias DisclosureAn independent demographic evaluation the vendor has submitted to, such as the NIST face evaluation class, or a governance framework with named process behind it.
Vendor Published

Before a model ships, Stripe tests it on authorization, false positive and block rates across all merchants and across subsets of them, and adjusts the model or policy for groups of merchants when a change would shift their outcomes badly. That is a named process for catching regressions across commerce segments. How block rates fall across the cardholders behind those payments is not reported.

AI Liability and Recourse
BB on AI Liability and RecourseA published falsifiable commitment such as an accuracy figure with its method, or a real correction route for the affected person, such as step up verification instead of silent denial.
Vendor Published

Platforms using Radar must let a connected account seek human review when Radar informed a decision about it, and must pass Stripe any information suggesting the output was wrong, which gives the affected business a real correction route. For payments, the merchant stays responsible for what it accepts, fraud disputes included. Chargeback Protection pays only on protected charges where Stripe has activated it, and standard screening carries no reimbursement for a wrong score.

Integration and Deployment
Model Supply Chain Disclosure
BB on Model Supply Chain DisclosureSubstantial partial disclosure, or a chain that is structurally short: an explicit in house build, on premise deployment, per customer instances, or zero retention at the model layer.
Vendor Published

Models are built and trained by Stripe on activity across its services, Financial Connections data, third party providers and public sources, and the Privacy Center separates the transaction fraud models behind Radar from the authorization models in Stripe Issuing. Radar runs specialized models for some payment methods, and Verifi and Ethoca supply dispute prevention signals. The third party data providers behind training, and the infrastructure that runs each plan's models, are not named.

Core Systems and Integration Depth
AA on Core Systems and Integration DepthNamed integrations with the systems of record, core banking, policy administration, custodial or contact center platforms, verifiable in marketplace listings or public API documentation.
Vendor Published

On Stripe, Radar attaches to Payments, Checkout, Billing and Connect through objects such as charges and payment intents and needs no extra integration. Businesses on another processor send payment, account and customer signals to the Radar API and act on the results in their own stack. Dispute prevention connects to Verifi and Ethoca, and Smart Disputes assembles and submits evidence on the merchant's behalf.

Deployment Model and Data Residency
CC on Deployment Model and Data ResidencyCloud only with nothing stated, which is the category norm.
Vendor Published

The service runs hosted inside Stripe's payments infrastructure and through an API for businesses that process elsewhere, with rules, reviews and analytics in the Dashboard. Card data sits in a separate, isolated AWS environment, and the data processing agreement covers international transfers. A residency commitment for Radar scoring and model data, a private cloud option or a version that runs in customer infrastructure is not stated.

Commercial
Commercial Transparency
AA on Commercial TransparencyPublished per unit rates a buyer can price against before any conversation.
Vendor Published

The entry tier, Radar Lite, is included with standard Stripe payments pricing. Standard, Plus and Pro cost $0.05, $0.07 and $0.09 per screened transaction pay as you go, with Pro adding $0.005 per abuse event, or start at $10, $14 and $20 a month with 200 screened transactions included and the same per transaction rate beyond. Platform plans start at $20, $44 and $70 a month, and enterprise access to Radar's signals for payments processed elsewhere is quoted.

Institution and Segment Coverage
AA on Institution and Segment CoverageThe financial segments served are named and each carries its own maintained material, whether the coverage is broad or deliberately narrow.
Vendor Published

Separate plans serve businesses that take payments and platforms and marketplaces with connected accounts, where Radar adds fraud scores for every account, automatic reserves and payout restrictions, and the service can score payments processed by another provider. FreshBooks screens businesses joining its payments service, DoorDash tested it on payments outside Stripe, and Anthropic uses it on its own payments. Subscription, usage based and travel businesses get abuse controls for trials, account sharing and pay as you go usage.

Alternatives to Stripe Radar

Other products that do the same kind of work as Stripe Radar, and what each one does.

Chargeflow automates chargeback prevention and recovery for ecommerce and SaaS merchants and for the payment platforms that serve them, with more than 20,000 merchants in its network across 90 countries.

Coris supplies merchant risk infrastructure to the parties that underwrite merchants rather than to merchants themselves, covering software platforms with embedded payments, independent sales organizations, payment facilitators, acquiring and sponsor banks, marketplaces and lenders.

Disputifier is a chargeback defense platform for ecommerce merchants, ISOs, CRMs, payment service providers and enterprises running thousands of merchant IDs.

Forter decides in real time whether to trust the person or AI agent behind an online purchase, login or return. Retailers, marketplaces and travel brands use it to approve more good orders and stop fraud, account takeover and policy abuse.

G2 Risk Solutions sells merchant risk intelligence to banks, merchant acquirers, payment providers and marketplaces, and since its combination with EverC, announced in August 2025, it sells the former EverC products for detecting high risk sellers, transaction laundering and illicit or…

NoFraud sells full service ecommerce fraud screening to small and mid sized merchants, with a Shopify centered distribution model and commercial terms a merchant can test without paying.

Commercial

Pricing

Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.

Entry Price Pricing Basis Data Protection Terms Implementation Source
Radar Lite has no added charge on standard Stripe payments pricing. Paid plans start at $10 per month.
$10 baseline
Radar Standard, Plus and Pro cost $0.05, $0.07 and $0.09 per screened transaction pay as you go, and Pro adds $0.005 per abuse event. Subscriptions start at $10, $14 and $20 a month with 200 screened transactions included and the same per transaction rate beyond. Platform plans start at $20, $44 and $70 a month. Enterprise pricing, including risk signals for payments processed elsewhere, is quoted. Stripe's Radar terms and data processing agreement cover use and sharing of data. The published plans do not list distinct data protection terms. Stripe offers an integration through its payments products and an interface for other processors. Enterprise integration support is quoted. Vendor Published

Radar Lite covers card fraud and card testing, with fraud alerts, at no added charge on standard Stripe payments pricing. Plus adds risk scores, custom rules with backtesting, manual reviews, risk preferences, Adaptive 3D Secure, Smart Refunds and Radar Assistant. Pro adds dynamic risk thresholds and abuse screening for free trials, bots, multiple accounts, account sharing and pay as you go usage.

Contact us

Found a vendor we missed? Have feedback on the index? We’d love to hear from you.

AI FinTech Index

The AI FinTech Index is an independent index that tracks changes to AI vendors in financial services. It holds 685 vendors across banking, lending, insurance, wealth, capital markets and financial crime compliance, each graded on the same 15 capability axes from public sources. No vendor pays for inclusion, placement, or rating.

Index Status
Last index update
November 20, 2026
The AI FinTech Index is an editorial reference, not a regulatory body. Vendor data is verified against published sources and public regulatory filings. Figures labeled “Estimated” have not been confirmed by the vendor. See the Methodology page for evaluation standards and limitations.
© 2026 AI FinTech Index
3801 N Capital of Texas Hwy, Ste E240 · Austin, TX 78746