DeepSeek Suffers Service Disruption, Affecting Website, App and APIs

China's artificial intelligence platform DeepSeek experienced a service disruption on the morning of October 15, 2026, leaving some users unable to access its chatbot normally. Users reported receiving a “server temporarily unavailable” message when attempting to start conversations, prompting widespread discussion on social media.

According to reports from Yicai and The Paper, DeepSeek's official service-status page recorded a partial outage affecting its web and API services at around 8:11 a.m. Beijing time. The affected components included the chat service and certain API functions.

Website, Mobile App and API Services Affected

Following the disruption, users reported difficulties accessing DeepSeek and completing chat requests. Some users said they could not start new conversations, while others experienced interruptions to their normal workflows.

The incident was not limited to a single access method. Information on the official status page indicated that both the web-based chat service and some API services were affected.

The disruption may have inconvenienced users who rely on DeepSeek for writing, coding assistance, research, document preparation and other daily tasks. For businesses and developers that incorporate AI tools into their workflows, even a temporary interruption can delay ongoing work.

Company Identifies the Cause and Begins Repairs

DeepSeek subsequently updated its official status page, stating that the cause had been identified and that a fix was being implemented.

However, media reports published later on October 15 indicated that some users were still unable to access the service normally. Full restoration had not yet been confirmed at the time of those reports.

The company had not released a detailed explanation of the technical cause, the full scope of the disruption or the expected restoration time. Online speculation about computing capacity, system upgrades or a possible new model release had not been substantiated and should not be treated as an established explanation.

Repeated Incidents Raise Questions About Reliability

Public service records show that DeepSeek has experienced other service disruptions and periods of degraded performance in recent weeks. On October 12, the platform recorded incidents involving partial interruptions to its chat service and reduced performance.

As AI tools become increasingly integrated into work, software development and education, reliability has become an important part of the user experience. Alongside model quality, response speed and cost, consistent service availability is essential for users who depend on AI systems for routine or business-critical tasks.

Repeated or prolonged outages can disrupt workflows and affect users' confidence in a platform's reliability.

Full Restoration Still Awaited

As of the morning of October 15, DeepSeek had said it was implementing a fix, but complete restoration had not been confirmed in the reports available at that time. Users should consult the company's latest service-status updates for confirmation that all affected functions have returned to normal.

The incident highlights the importance of reliable infrastructure and effective incident response as AI services attract broader adoption. Further information about the cause of the disruption, the recovery process and any subsequent technical improvements remains pending.

Sources:Yicai, The Paper, Sina Tech and DeepSeek's official service-status page, October 15, 2026.