Anthropic's Claude suffers partial outage on September 29

By Splitfeed EditorialLast updated Sep 29, 2026, 16:20 UTC

On Tuesday, September 29, 2026, Anthropic's Claude AI chatbot experienced a partial outage that affected thousands of users in the United States and internationally. Downdetector tracked a surge of reports indicating that users could not reach the service through either the app or the web. The disruption was partial, meaning some users retained access while others were locked out entirely. Spotify also went down during the same window, though no connection between the two outages was established. Benzinga later reported that Claude had returned to service before the end of the day, and no official cause for the disruption was identified.

The outage matters beyond the headline because Claude has grown into a widely used commercial AI assistant, with business and developer workflows increasingly dependent on its availability. Any reliability gap at this stage of AI adoption raises practical questions about infrastructure resilience for services that position themselves as productivity tools. The coincidence with Spotify's outage is a detail that attracted attention but remains unexplained, and the absence of any stated cause from Anthropic leaves open questions about what triggered the disruption and how quickly the company can diagnose and communicate such events in the future.

The anti-AI camp did react, with Jani Yli-Paavola posting the Finnish phrase "Mut ei tässä mitään kuplaa ole," which translates roughly to "But there's no bubble here," a sardonic remark suggesting the outage fits a broader skepticism about inflated confidence in AI services. The pro-AI camp has produced no published reactions to this story yet. Typically that camp would frame a brief, resolved outage as routine growing pains for a rapidly scaling service, comparable to disruptions at any major cloud platform. The middle ground camp has also not weighed in, but stories like this generally prompt that group to call for clearer uptime standards and more transparent incident communication from AI providers.

The clearest thing to watch going forward is whether Anthropic publishes a post-incident report explaining the cause and duration of the outage, and whether the frequency of such disruptions increases or decreases as the service scales heading into late 2026.