A later update described degraded performance across a wider group of products, including Claude.ai, the Claude API, Claude Code, Claude Cowork, and Claude Console.

The confirmed cause—and what remains unknown

The incident began as an authentication failure: some users could not log in or access Claude. Anthropic later said it had identified the cause and deployed a fix.

However, the available reports do not identify the underlying technical fault, such as a particular identity provider, deployment, database, or network component. It is therefore more accurate to call this an authentication-related outage than to claim a specific root cause that Anthropic had not publicly detailed.

Timeline: start, expansion, and resolution

• 21:58 UTC: Anthropic began investigating authentication problems affecting Claude.ai, Claude Code, and Claude Cowork.
• 22:02 UTC: The reported scope widened to include degraded performance across additional Claude services.
• 22:34 UTC: Anthropic marked the incident resolved, putting the total duration at approximately 36 minutes.

Third-party outage reports peaked at more than 15,000, indicating that the disruption was visible to a large number of users even though recovery came relatively quickly.

Which Claude services were affected?

The available coverage describes the incident in two stages:

• Initially affected: Claude.ai, Claude Code, and Claude Cowork.
• Included in later reports of degraded performance: Claude.ai, Claude Code, Claude Cowork, the Claude API, and Claude Console.

The difference matters because an authentication problem can appear first in user-facing products and then be reported more broadly as engineers assess related platform components. The sources provided do not offer a definitive list of products that remained fully unaffected during the entire incident. The initial report named three services; later coverage reported degradation across five.

How unusual was this outage in 2026?

The August 16 incident came after several other Claude disruptions reported during 2026, but those events were not all attributed to the same cause.

Repeated disruptions in June

TechTimes reported that Claude experienced its tenth significant disruption since June 5 by mid-June. That coverage linked the broader reliability concern to demand growing faster than Anthropic’s infrastructure could handle.

That explanation should be treated as context for the period, not as a confirmed explanation for the August 16 authentication incident. The available August 16 reporting says Anthropic identified and fixed the cause but does not provide a detailed technical postmortem.

Multiple incidents in late July

Coverage of July 29–30 described two separate incidents within 24 hours affecting Claude.ai, the API, Claude Code, and Claude Cowork. One account said multiple network failures reduced Anthropic’s capacity to serve Claude while traffic was rerouted.

This was a different incident pattern from August 16’s reported authentication failure, although both affected several products at once.

A lengthy model-related outage in August

A separate August 5 disruption affected Claude Mythos 5, Fable 5, Opus 5, and Sonnet 5. TechTimes reported that the combined incident lasted about seven hours and 29 minutes, including a later, separate degradation affecting Opus 5.

The August 5 event involved elevated model errors, while the August 16 event began with login and authentication failures. The available sources do not establish that the two incidents shared a root cause.

A wider reliability problem across AI services

Data cited by IEEE ComSoc from an Ookla reliability report found that high-disruption days across four major AI applications increased from 6 in Q1 2025 to 51 in Q1 2026. The same report attributed 39 of the 51 Q1 2026 disruption days to Claude.

Those figures provide industry context, but they do not prove why any individual Claude outage occurred. For the August 16 incident, the strongest supported conclusion is narrower: an authentication problem expanded into multi-product degradation and was resolved in roughly 36 minutes.

What users should take from the incident

The outage demonstrated how a failure at the authentication layer can affect more than the main Claude chat interface. Developers using Claude Code or the API, and teams relying on Claude Cowork or Console, could encounter problems at the same time.

It also reinforces the importance of checking Anthropic’s status page before repeatedly refreshing an account or changing an application integration. The incident was resolved quickly, but the wider 2026 record shows that Claude disruptions have varied in duration, scope, and reported cause.

The evidence supports describing August 16 as a brief, broad authentication-related outage—not as proof of one single infrastructure problem behind every Claude disruption in 2026.

A later update described degraded performance across a wider group of products, including Claude.ai, the Claude API, Claude Code, Claude Cowork, and Claude Console.

The confirmed cause—and what remains unknown

The incident began as an authentication failure: some users could not log in or access Claude. Anthropic later said it had identified the cause and deployed a fix.

However, the available reports do not identify the underlying technical fault, such as a particular identity provider, deployment, database, or network component. It is therefore more accurate to call this an authentication-related outage than to claim a specific root cause that Anthropic had not publicly detailed.

Timeline: start, expansion, and resolution

• 21:58 UTC: Anthropic began investigating authentication problems affecting Claude.ai, Claude Code, and Claude Cowork.
• 22:02 UTC: The reported scope widened to include degraded performance across additional Claude services.
• 22:34 UTC: Anthropic marked the incident resolved, putting the total duration at approximately 36 minutes.

Third-party outage reports peaked at more than 15,000, indicating that the disruption was visible to a large number of users even though recovery came relatively quickly.

Which Claude services were affected?

The available coverage describes the incident in two stages:

• Initially affected: Claude.ai, Claude Code, and Claude Cowork.
• Included in later reports of degraded performance: Claude.ai, Claude Code, Claude Cowork, the Claude API, and Claude Console.

The difference matters because an authentication problem can appear first in user-facing products and then be reported more broadly as engineers assess related platform components. The sources provided do not offer a definitive list of products that remained fully unaffected during the entire incident. The initial report named three services; later coverage reported degradation across five.

How unusual was this outage in 2026?

The August 16 incident came after several other Claude disruptions reported during 2026, but those events were not all attributed to the same cause.

Repeated disruptions in June

TechTimes reported that Claude experienced its tenth significant disruption since June 5 by mid-June. That coverage linked the broader reliability concern to demand growing faster than Anthropic’s infrastructure could handle.

That explanation should be treated as context for the period, not as a confirmed explanation for the August 16 authentication incident. The available August 16 reporting says Anthropic identified and fixed the cause but does not provide a detailed technical postmortem.

Multiple incidents in late July

Coverage of July 29–30 described two separate incidents within 24 hours affecting Claude.ai, the API, Claude Code, and Claude Cowork. One account said multiple network failures reduced Anthropic’s capacity to serve Claude while traffic was rerouted.

This was a different incident pattern from August 16’s reported authentication failure, although both affected several products at once.

A lengthy model-related outage in August

A separate August 5 disruption affected Claude Mythos 5, Fable 5, Opus 5, and Sonnet 5. TechTimes reported that the combined incident lasted about seven hours and 29 minutes, including a later, separate degradation affecting Opus 5.

The August 5 event involved elevated model errors, while the August 16 event began with login and authentication failures. The available sources do not establish that the two incidents shared a root cause.

A wider reliability problem across AI services

Data cited by IEEE ComSoc from an Ookla reliability report found that high-disruption days across four major AI applications increased from 6 in Q1 2025 to 51 in Q1 2026. The same report attributed 39 of the 51 Q1 2026 disruption days to Claude.

Those figures provide industry context, but they do not prove why any individual Claude outage occurred. For the August 16 incident, the strongest supported conclusion is narrower: an authentication problem expanded into multi-product degradation and was resolved in roughly 36 minutes.

What users should take from the incident

The outage demonstrated how a failure at the authentication layer can affect more than the main Claude chat interface. Developers using Claude Code or the API, and teams relying on Claude Cowork or Console, could encounter problems at the same time.

It also reinforces the importance of checking Anthropic’s status page before repeatedly refreshing an account or changing an application integration. The incident was resolved quickly, but the wider 2026 record shows that Claude disruptions have varied in duration, scope, and reported cause.

The evidence supports describing August 16 as a brief, broad authentication-related outage—not as proof of one single infrastructure problem behind every Claude disruption in 2026.