Right now, “what happened to Claude” mostly refers to a big outage and reliability drama around Anthropic’s Claude AI, not to the company shutting it down or anything similar.

Quick Scoop

  • Claude had a major outage on March 2, 2026, with lots of users suddenly unable to log in or get replies.
  • The issue was described as “elevated errors” affecting claude.ai, login/logout, Claude Opus 4.6, and dev tools like Claude Code and Claude Console.
  • Anthropic said the disruption was caused by surging demand / unprecedented load after Claude shot to the top of the US App Store downloads.
  • Service has since been restored , and Claude is still active and being updated (latest flagship model line includes Claude Opus 4.6).

What exactly went wrong?

1. The March 2026 outage

Reports from tech outlets describe:

  • A “significant outage” lasting over two hours on a Monday morning.
  • Errors across:
    • claude.ai (the chat site)
    • Claude Code (coding assistant)
    • Claude Opus 4.6 (latest top model)
  • Status pages and company updates saying they were:
    • Investigating login / logout failures
    • Dealing with malfunctioning API methods
    • Handling “unprecedented demand” and “elevated errors.”

Anthropic later stated that Claude was “back and running” and thanked users for their patience, emphasizing that the rush of new users had stressed their infrastructure.

2. Why everyone is asking “what happened to Claude?”

A few overlapping things have made the phrase trend:

  • Outage + hype spike
    • Claude recently became one of the most downloaded free apps on the US App Store, even overtaking some very big rivals.
* That surge in new users appears to have pushed infrastructure over the edge, which turned into a visible outage just as attention was highest.
  • Political / government angle
    • Coverage mentions a dispute where the US government decided to stop using Anthropic tech (including Claude), and another vendor stepped in for certain defense/“classified environment” use cases.
* That added a “drama” layer to what would otherwise be a plain scaling incident.
  • Past reliability drama
    • In 2025, there were already technical postmortems about Claude’s degraded responses : strange characters suddenly appearing, routing bugs, and token-selection issues that made answers feel off, even when the system was “up.”
* Those earlier issues were eventually fixed, but they left a narrative that “Claude sometimes breaks in weird ways.”

Put together: outages + older bugs + government/political news + app-store virality = lots of users asking “what happened to Claude?”

Is Claude dead, replaced, or shut down?

No:

  • Anthropic is still actively shipping new Claude models, including Claude Opus 4.6, with improved coding, long-running task management, and professional-quality output.
  • The March 2026 disruption was a temporary service outage , not a shutdown of the product.
  • After the outage window, Anthropic said Claude is operational again and serving users.

So “what happened” is best understood as:

Claude is still around and evolving, but it hit a patch of instability: earlier quality glitches in 2025, and a big demand-driven outage in early March 2026 that made lots of people notice and talk about it.

Mini timeline

  1. 2025 – Quality / reliability issues
    • Users report odd outputs (random Thai/Chinese characters, degraded wording) and routing bugs; Anthropic publishes technical explanations and fixes.
  1. Late 2025 – early 2026 – Capability push
    • New Claude 4.x models, including Opus 4.6, focus on stronger coding, planning, research, and long-context work, mostly for professional and enterprise use.
  1. Early 2026 – Popularity + outage
    • Claude climbs to the top of the US App Store free charts; downloads and free-user numbers surge.
 * Infrastructure strains under “unprecedented demand,” causing outages and login/API issues in early March 2026.

 * Service is brought back online; Anthropic frames it as a scaling issue that they are working to stabilize.

Multiple viewpoints

  • Optimistic take
    • The outage is a growing pain for a rapidly scaling product whose demand exceeded expectations.
* The earlier detailed postmortems show a willingness to expose technical problems and fix them.
  • Critical take
    • Repeated reliability incidents (2025 glitches + 2026 outage) raise questions about robustness under load, especially for people using Claude for serious work.
* The loss of a US government relationship in favor of a competitor suggests that, at least in some high-stakes environments, confidence may have shifted.
  • Pragmatic user take
    • For everyday users, it mostly means: you might occasionally see downtime or weird behavior during periods of heavy demand, but the service is not gone and continues to be developed.

TL;DR: Claude hasn’t disappeared; it hit infrastructure and reliability issues—most notably a big demand-driven outage in March 2026—while at the same time growing fast in popularity and continuing to release new, more capable versions.

Information gathered from public forums or data available on the internet and portrayed here.