⚠ CLAUDE.AI OUTAGE HISTORY — 965+ INCIDENTS SINCE JUNE 2024     137 INCIDENTS IN 90 DAYS     43 MAJOR OUTAGES     NO SLA     NO AUTOMATIC REFUNDS     ⚠ CLAUDE.AI OUTAGE HISTORY — 965+ INCIDENTS SINCE JUNE 2024    
claudesucks.workwithvsg.com / claude-ai-outages
⚠ Outage Record — Last Updated April 28, 2026

Claude.AiOutage History.

The complete documented record of Claude.ai service failures, downtime, and outages sourced from Anthropic's own status page, StatusGator, IsDown, and Downdetector. 965+ incidents since June 2024. This page is updated as new outages are confirmed.

965+
Total incidents
since June 2024
137
Incidents in
last 90 days
43
Major outages
last 90 days
68min
Median outage
duration
$0
Auto refund
for subscribers
// april 2026

Recent Outages

The most recent documented outages affecting Claude.ai subscribers. Sources: Anthropic status.claude.com, StatusGator, IsDown, Downdetector.

Claude.Ai Unavailable — Elevated Api Errors
MAJOR APR 28, 2026 1HR 24MIN
Claude.ai completely unavailable alongside elevated errors on the Anthropic API and login failures for Claude Code. Over 12,000 user reports logged on Downdetector at peak. Anthropic confirmed the outage as Major. No advance notice to subscribers.
Official Timeline
17:34 UTCImpact begins — users unable to reach Claude.ai
17:41 UTCAnthropic begins investigating
17:51 UTCIssue identified: elevated errors on API and Claude.ai login paths
18:52 UTCIncident resolved — 1 hour 18 minutes of confirmed downtime
Source: status.claude.com, StatusGator, Downdetector, Tom's Guide live blog April 28 2026
Elevated Errors — Claude Sonnet 4.5
WARNING APR 28, 2026 19MIN
Elevated error rates on Claude Sonnet 4.5 affecting API responses. Separate from the major outage later the same day.
Source: status.claude.com, StatusGator
Service Degradation — 18 Hour Warning
WARNING APR 25, 2026 18HR 43MIN
Extended degraded service warning lasting over 18 hours. Chats failing to load, slower responses globally across North America, Europe, and Asia-Pacific. Claude's status page showed ongoing investigation with engineers triaging elevated errors across multiple regions.
Source: StatusGator outage history, status.claude.com
Extended Warning Period
WARNING APR 24, 2026 13HR 50MIN
13+ hours of degraded service followed by a 39-minute full outage the same day.
Source: StatusGator
Claude.Ai Down — Extended Outage
MAJOR APR 20, 2026 6HR 50MIN
Nearly 7 hours of confirmed downtime affecting Claude.ai. One of the longest single outages in April 2026. No proactive notification to subscribers.
Source: StatusGator
Mcp Applications Unavailable
WARNING APR 23, 2026 15HR 19MIN
MCP applications across Claude.ai including desktop and mobile unavailable for over 15 hours. Structured outputs also affected during this period returning reduced quality responses.
Source: status.claude.com, StatusGator
Claude.Ai And Platform Down — Login Failures
MAJOR APR 15–16, 2026 ~3HR
Cascading failures beginning with elevated API errors, escalating to full platform outage. Login for Claude Code broken. Pro users unable to authenticate. Free users falsely told they had exhausted their daily message limit. Anthropic confirmed at peak: "Claude.ai and Platform are down."
Official Timeline
Apr 15 – 10:53 ETElevated errors confirmed on Claude.ai, API, Claude Code
Apr 15 – 11:03 ETFix attempted
Apr 15 – 11:40 ET"Claude.ai and Platform are down" — full outage confirmed
Apr 15 – 13:42 ETIncident resolved
Apr 16 – 06:03 PTOpus 4.6 outage begins
Apr 16 – 07:26 PTOpus 4.6 restored
Source: TechRadar live blog April 15 2026, status.claude.com
// march 2026

March 2026:
The Worst Month On Record

March 2026 saw Claude's most severe and repeated outages. Anthropic cited "unprecedented demand" as a contributing factor. Users experienced cascading failures across multiple components on multiple occasions within days of each other.

Multi-Wave Global Outage — Worst Incident On Record
MAJOR MAR 2–3, 2026 ~20HR TOTAL
The most disruptive outage in Claude's documented history. Three separate waves of failure across the same incident window. HTTP 500 and 529 errors, login failures, complete timeouts. Between 1,700 and 4,700 Downdetector reports during the first wave alone. Anthropic attributed it in part to unprecedented demand. Developers saw elevated failure rates across all API endpoints.
Official Timeline (Condensed)
Mar 2 – 13:37Issue identified — fix being implemented
Mar 2 – 14:35Elevated errors on claude.ai, console, Claude Code confirmed
Mar 2 – 15:47First wave resolved
Mar 2 – 17:15Second wave begins
Mar 2 – 18:18Issue repeats — third investigation
Mar 2 – 21:16Second wave resolved
Mar 3 – 03:15Third wave begins
Mar 3 – 09:36Fix implemented — monitoring begins
Source: Data Business Central, status.claude.com, Downdetector
Global Disruption — Api, Platform, Claude Code, Cowork
MAJOR MAR 25, 2026 ~1HR
Sign-in failures, timeouts, and elevated errors across Claude API, platform.claude.com, Claude Code, and Cowork simultaneously. Global impact — users from North America, Europe, India, Japan, Canada, Germany, Indonesia affected. StatusGator detected the issue 1 hour before Anthropic acknowledged it.
Source: StatusGator outage history
Claude.Ai Chat Outage — Complete Failure
MAJOR MAR 19, 2026 ~70MIN
Escalated from degraded to complete outage within 5 minutes of initial detection. Services returned to normal after approximately 70 minutes.
Source: StatusGator
Elevated Errors — Chat And Desktop
WARNING MAR 23, 2026 16MIN
Elevated errors on Claude.ai while API remained available. StatusGator detected the issue 49 minutes before Anthropic acknowledged it on the status page.
Source: StatusGator
// february 2026

February 2026

Usage Reporting Outage — Data Loss
MAJOR FEB 26–27, 2026 ~28HR
Usage reporting via Console and admin API failed. Approximately 28 hours of missing usage data. All historical data after Feb 25 18:40 UTC was unavailable until eventual restoration. Enterprise subscribers with billing based on usage reporting were directly affected.
Official Timeline
Feb 26 – 15:56Outage identified in usage reporting
Feb 26 – 18:52Partial fix deployed
Feb 27 – 03:04Reporting restored but data after Feb 25 18:40 UTC missing
Feb 27 – 13:12Fully resolved — historical data restored
Source: status.claude.com via Data Business Central
Platform Error Surge — 4,700 User Reports
MAJOR FEB 24, 2026 UNCONFIRMED
Over 4,700 user error reports. HTTP 500 internal server errors and broad communication failures across Claude chatbot and developer tools. Raised questions about infrastructure capacity ahead of the even larger March outages that followed days later.
Source: Data Business Central, Downdetector
// november 2025

November 2025

Cloudflare Cascade — Global Ai Platforms Down
MAJOR NOV 18, 2025 EXTENDED
A widespread Cloudflare infrastructure failure took down Claude alongside ChatGPT and other major AI platforms globally. Users worldwide encountered error messages and latency spikes affecting both consumer chat and business API access. This incident exposed a critical systemic vulnerability: Claude's reliability is dependent on third-party infrastructure it does not control. Subscribers paid for a service that was unavailable through no fault of their own use — and received no automatic compensation.
Source: Data Business Central, StatusGator
// the full record

By The Numbers:
The Complete Picture

These are not anomalies. This is the product as it operates. Every figure below is sourced from third-party monitoring services using Anthropic's own official status page as their primary data source.

965+
Total incidents recorded since June 2024 by StatusGator monitoring Anthropic's own status page
2+
Average incidents per day across the platform since tracking began
87
Incidents in March 2026 alone that Anthropic never officially reported on their status page
33
Outages in March 2026 detected by third-party monitors up to 2.3 hours before Anthropic acknowledged them
5
Separate components tracked: claude.ai, Claude API, platform console, Claude Code, Cowork — each can fail independently
$0
Automatic compensation issued to any subscriber for any outage at any point in Claude's history

Demand Your
Refund.

Anthropic will not contact you about downtime. You must file the complaint yourself. Here is where to send it.

// file a complaint

Submit to Anthropic's consumer safety address with your outage dates, subscription type, and a refund request. Demand written confirmation.

usersafety@anthropic.com support@anthropic.com

// legal escalation

Jeffrey Bleich, General Counsel of Anthropic as of August 2025. Certified mail creates a legal record email does not.

Anthropic PBC, 548 Market St
PMB 90375, San Francisco CA 94104
Attn: Jeffrey Bleich, General Counsel

// add your experience

Document your outage on the main site. Every submission adds to the public record available to other subscribers and attorneys.

claudesucks.workwithvsg.com
// more on claudesucks
CONSUMER WARNING — HOME
The full case against Claude.ai including eleven documented failures, on-record admissions, and withheld truths.
FALSE CAPABILITY CLAIMS
Every false capability claim made in a single documented session, confirmed on the record by Claude itself.
HOW TO GET YOUR REFUND
Step-by-step guide to demanding compensation from Anthropic for outages and failed sessions.