CLAUDE LABJP
FORK — Claude Code 2.1.212 changes what /fork does: it copies your conversation into a new background session with its own row in claude agents, so you can keep working. The old in-session subagent is now /subtaskLIMITS — WebSearch calls are now capped at 200 per session by default, and subagent spawns get the same 200 ceiling, so a runaway search or delegation loop stops on its ownMCPBG — MCP tool calls running past two minutes now move to the background automatically, keeping the session usable. Tune the threshold with CLAUDE_CODE_MCP_AUTO_BACKGROUND_MSPLANFIX — Fixed plan mode auto-running file-modifying Bash commands such as touch and rm without a permission prompt or an SDK canUseTool callbackSONNET5 — Claude Sonnet 5 is running on introductory pricing of $2 per million input tokens and $10 per million output. After August 31 it moves to $3 and $15IPO — Bankers are reportedly lining up investor meetings for Anthropic ahead of a possible public listing as soon as OctoberFORK — Claude Code 2.1.212 changes what /fork does: it copies your conversation into a new background session with its own row in claude agents, so you can keep working. The old in-session subagent is now /subtaskLIMITS — WebSearch calls are now capped at 200 per session by default, and subagent spawns get the same 200 ceiling, so a runaway search or delegation loop stops on its ownMCPBG — MCP tool calls running past two minutes now move to the background automatically, keeping the session usable. Tune the threshold with CLAUDE_CODE_MCP_AUTO_BACKGROUND_MSPLANFIX — Fixed plan mode auto-running file-modifying Bash commands such as touch and rm without a permission prompt or an SDK canUseTool callbackSONNET5 — Claude Sonnet 5 is running on introductory pricing of $2 per million input tokens and $10 per million output. After August 31 it moves to $3 and $15IPO — Bankers are reportedly lining up investor meetings for Anthropic ahead of a possible public listing as soon as October
Articles/Claude.ai
Claude.ai/2026-07-09Intermediate

Claude Card Declined: A Complete Troubleshooting Guide for Pro, Max, and API Users

When Claude tells you 'Your card was declined,' the cause is rarely obvious from the error text alone. This guide separates the three layers where a decline actually happens — your issuing bank, Stripe's fraud engine, and Anthropic's account state — and walks you through fixes plus fallback payment methods that almost always get the charge through.

Claude46PaymentCard DeclinedStripe15Troubleshooting11Anthropic13International Payments

If You're Stuck on 'Your Card Was Declined'

You tried to subscribe to Claude Pro and saw "Your card was declined." You tried to add API credits in the console and got "Payment failed." These two messages from Anthropic look like they should tell you what went wrong, but they don't — identical text fires for dozens of unrelated root causes, from a mistyped CVV to a fraud model on the other side of the world flagging your transaction.

I've hit this myself across personal credit cards, a corporate card, and a debit card, and what I learned is that the problem is almost never with Anthropic. A decline is happening in one of three layers: your issuing bank, Stripe (Anthropic's payment processor), or your Anthropic account itself. Knowing which layer owns the failure is most of the battle.

This guide walks through how to figure out which layer is blocking you, what to do at each one, and — if none of that works — which fallback payment method to reach for. In most cases you can get a payment through in under five minutes once you know where to look.

Read the Error Message Carefully First

Anthropic's decline messages reduce to two real categories, regardless of the exact wording.

The first is "Your card was declined" / card_declined. Stripe asked your bank to authorize the charge, and your bank said no. Anthropic can do nothing about this — the decision was made at the bank. You need to check your banking app or call the number on the back of the card.

The second is "Payment failed" / "Your payment could not be completed". This one was blocked by Stripe or Anthropic before it ever reached the bank. Stripe's Radar fraud engine flagged the attempt as risky, or the billing address doesn't match the card country, or Anthropic has placed a hold on your account.

English error codes are more useful than the Japanese translations. card_declined is the first category. do_not_honor, fraudulent, and processing_error usually mean the second. Screenshot the exact message and code before you start troubleshooting — you'll want it if you have to contact support.

The Three-Layer Diagnostic Checklist

Decline causes roughly follow this distribution, in order of likelihood:

Layer 1: Your issuing bank (~60%) Most declines are your bank blocking international online subscriptions. Cards issued by Japanese banks, UK high-street banks, and several EU retail banks aggressively flag first-time charges to US-registered merchants, and Anthropic is a US entity.

Layer 2: Stripe Radar (~25%) Stripe runs a machine-learning fraud engine called Radar. It flags transactions where the billing country and IP country don't match, where the same card has been retried too many times in a short window, or where the browser fingerprint matches recent failed attempts. Once Radar flags you, even a good card won't go through for an hour or two.

Layer 3: Anthropic account state (~15%) Often overlooked. If you burned through your free API trial, if you have an unpaid invoice from a separate account, or if the account is under review for terms-of-service issues, Anthropic won't accept the payment regardless of the card.

Work through these in order of ease: check your bank app for recent blocks → try a different browser and IP → try a different card. Do not keep retrying the same card rapidly — that actively makes layer 2 worse.

Layer 1 Fix: Unblocking at the Issuing Bank

Most "international online payment blocks" can be toggled in your bank's app.

For the major Japanese card issuers: Mitsui Sumitomo (Vpass) has it under "Card Usage Restrictions." JCB (MyJCB) has it under "International Online Fraud Prevention." Rakuten (e-NAVI) has it under "Overseas Merchant Usage Setting." American Express doesn't block international online payments by default, so if an Amex is declining, it's almost certainly a layer 2 issue.

Wait 15–30 minutes after changing the setting before you retry. Banks take time to propagate changes. Retrying immediately and getting declined again can escalate you to layer 2, which is worse.

Banks with app-based approval (Kyash, Revolut, Wise, Sony Bank WALLET, and increasingly N26 and Chase in the US) are the easiest to use — when Anthropic tries to charge, you get a push notification and approve it with one tap. This path leaves no Radar footprint, so it's my top recommendation for first-time Claude subscriptions.

Layer 2 Fix: Getting Past Stripe Radar

You can't call Stripe. You can only give Radar less reason to be suspicious.

Three things move the needle.

First, make the billing country match the IP country. If you're in Japan using a Japanese card, turn off your VPN and check out from a Japanese IP. A US IP with a JP card triggers Radar almost every time. The opposite also holds.

Second, clear the browser state completely. Stripe fingerprints the browser, so retrying from the same Chrome window with failed attempts already in its history keeps failing. Open an incognito window or switch to Safari/Firefox, clear cookies for stripe.com and anthropic.com, and then enter the card fresh.

Third, wait. Three consecutive declines on the same card puts it in a 24-hour cooldown where nothing works. Come back tomorrow — nearly everyone who posts "it randomly worked the next day" hit this cooldown.

Layer 3 Fix: Anthropic Account Issues

Log in at console.anthropic.com and open Settings → Billing. If "Add payment method" isn't there, or you see "Account is locked," only Anthropic support (support@anthropic.com) can unblock you. Expect 48–72 hours for a response.

Patterns that need support help:

  • You exhausted the $5 free trial and never attached a payment method
  • A separate Anthropic account tied to the same email has an unpaid invoice
  • The account was flagged for account-sharing or suspected abuse

Also worth knowing: Claude Pro (claude.ai) and the API (console.anthropic.com) have separate billing surfaces even when the email is the same. Updating the card on Pro does not update it on the API side. If you're using both, update both.

Fallback 1: Wise or Revolut Multi-Currency Cards

When a local credit card won't go through no matter what, my strongest recommendation is Wise. You fund a Wise balance in your home currency, convert to USD inside the app (around 0.4% fee), and then pay Anthropic from the USD balance. From Stripe's perspective this looks like a domestic US dollar charge — Radar risk drops through the floor. In my experience, decline rates after switching to Wise are under 5%.

The flow is:

  1. Open a Wise account (identity verification takes 2–3 days)
  2. Fund your Wise account from your domestic bank
  3. Convert the funds to USD inside Wise
  4. Register the Wise debit card on Anthropic

Revolut works similarly in regions where it's available. Wise is generally more reliable for Japanese residents in 2026 since Revolut's Japan service was scaled back.

For US-incorporated businesses, Brex and Mercury are the strongest options, but they require a US entity, which isn't practical for most individual international users.

Fallback 2: Corporate Cards and Apple Pay

If you're paying for Claude as a business expense, a corporate card almost always works. Japanese corporate cards (Mitsui Sumitomo Platinum Corporate, Amex Business, JCB Biz) ship with international online charges enabled by default and rarely trip Radar.

One caveat: putting Claude Pro (individual plan) on a corporate card creates accounting friction later — was it personal or business use? If the intent is business, contract through Claude for Work instead of using a personal account. Invoicing and seat management become much cleaner.

Apple Pay is an underrated fallback for Claude Pro. The Pro checkout supports Apple Pay, and because Apple Pay issues a device-specific token rather than revealing the real PAN, it can bypass some issuing-bank blocks. Note: the API side (console.anthropic.com) does not accept Apple Pay — only Pro does.

Three Pitfalls That Caught Me Personally

A few things the docs won't tell you.

Virtual cards get shut down mid-subscription. Kyash, Revolut's burner cards, and bank-issued one-time cards sometimes work on the first charge, then the issuer flags "recurring foreign merchant" as fraud and kills the card silently. You'll find out when Claude Pro fails to renew. Use a real card or a Wise debit for subscriptions, not a disposable virtual number.

3-D Secure is quietly mandatory now. Since late 2025, Stripe has enforced 3-D Secure (SCA) on most EU and Japanese cards. A card that passes the first charge can get declined on renewal if 3-D Secure was never set up on the card. Register the 3-D Secure password on your issuer's website before attaching the card to Anthropic.

Don't mix currencies. If Anthropic bills you in USD but your Wise card pays in JPY, the USD→JPY→USD round-trip looks like an FX arbitrage to Radar and can trigger a soft decline. Keep one currency end to end — if you're on USD billing, pay with a USD balance.

Retry Timing, and the Damage Caused by Retrying Too Often

Retrying can itself become the reason your card is declined. If you attempt the same card four or five times within a short window, Stripe Radar raises its risk score for suspected card testing — and from that point on, even perfectly valid card details stop going through.

As an indie developer who handles every billing problem alone, I keep one rule: after two consecutive declines, I stop. I wait at least thirty minutes, and I use that time to call the issuing bank or prepare a different payment method. Hitting submit a third time right away was consistently the single worst thing I could do.

It also helps to keep the environment stable across attempts. Retrying through a VPN or an incognito window tends to create a mismatch between the card's country of issue and the IP country code, which both tightens Radar's judgement and makes the underlying cause harder to isolate. For payment attempts, go back to your everyday browser on your everyday connection.

SituationWhat to do next
First declineRecheck the number, expiry, and billing address, then retry exactly once
Second declineStop. Check the authorization history in your banking app
After 30 minutesUnlock with your bank, or switch to a different card or Apple Pay

For behavior specific to Japanese issuers, "Why Rakuten Card Gets Declined on Claude" covers a separate diagnostic path, and if the failure happens when buying API credits, "Claude Pro/Max Billing and Payment Errors" walks through that side of the system.

Last Resorts When Nothing Works

If none of the above works, try these in order:

  • Email Anthropic support (support@anthropic.com) for an account state review. 48–72 hour turnaround
  • Switch to Claude for Work, which supports invoice / ACH billing and skips card processing entirely
  • Use Claude through Amazon Bedrock. Charges go to your AWS bill, which bypasses Anthropic's payment system
  • Use Claude through Google Cloud Vertex AI. Same story, via GCP

Bedrock and Vertex access are explicitly supported by Anthropic's official tools including Claude Code. If personal payment is a persistent problem, routing through a cloud provider you already pay is often the cleanest long-term fix.


A Claude card decline is almost always solvable once you know which layer to look at. Start at the bank — most cases resolve there. When they don't, Wise, Apple Pay, or a Bedrock/Vertex route will get you back to writing code instead of arguing with checkout forms.

Share

Thank You for Reading

Claude Lab is ad-free, supported entirely by members like you. We publish practical guides daily with implementation code, benchmarks, and production-ready patterns. If you've found it useful, we'd love to have you on board.

  • Copy-paste ready implementation code
  • New advanced guides published daily
  • $5/mo or $10 for lifetime access
View Membership →

If you found this article helpful, a small tip ($1.50) would mean a lot to us. Your support helps keep this site ad-free and covers server and hosting costs.

Related Articles

Claude.ai2026-05-04
Anthropic IPO Outlook and the Indie Developer Revenue Strategy 2026
Anthropic's IPO preparation phase is becoming concrete in 2026. For indie developers and freelancers building on Claude, here's how to think about pricing, dependency, and unique value layers — together with the structural changes that often arrive when a platform goes public.
Claude.ai2026-05-02
Anthropic IPO 2026 Status — Timing, Valuation, and What It Means for Claude Users
A grounded look at where Anthropic stands on going public as of May 2026. Covers timing speculation, valuation history, comparison with OpenAI, and what an eventual IPO might mean for Claude users.
Claude.ai2026-04-27
Anthropic IPO 2026 — A Builder's Roadmap in 7 Practical Lenses
Most Anthropic IPO coverage is written for investors. This piece reframes the same news through the eyes of a solo developer shipping with Claude — what it means for API pricing, model cadence, and the contract terms you should pay attention to.
📚RECOMMENDED BOOKS
Build a Large Language Model (From Scratch)
Sebastian Raschka
LLM Dev
Prompt Engineering for LLMs
Berryman & Ziegler
Prompting
AI Engineering
Chip Huyen
AI Eng
* Contains affiliate links
See all →