China’s DeepSeek suffered the biggest outage in its history as the popular AI chatbot went down for more than seven hours overnight in China. As per a Bloomberg report, the outage forced DeepSeek to deploy multiple updates to rectify the situation.
When did the DeepSeek outage happen?
According to outage tracking platform Downdetector, users first began reporting faults with the platform on Sunday evening. The startup’s official status page acknowledged an initial issue at 9:35 p.m., eventually marking the incident as resolved about two hours later.
However, the disruption continued into the following day. Subsequent updates on Monday showed that DeepSeek was addressing another case of performance issues, which took until 10:33 a.m. to be fully fixed.
What caused the extended downtime?
The exact causes of the massive outage currently remain unclear, and DeepSeek has not given an official statement regarding the reasons for the disruption.
Notably, DeepSeek, in its own status page, has maintained that it has a near 99% operational record since it first unveiled the R1 model in January 2025, making the outage an even more unusual event for the Chinese AI startup.
What’s happening with DeepSeek?
DeepSeek had its viral moment in January last year, when its AI models gained widespread attention and rattled Silicon Valley, triggering a sell-off in tech stocks and wiping off billions of dollars in wealth as investors began questioning assumptions around American dominance in the AI race.
Since then, however, it has been a long wait for DeepSeek fans. The startup is yet to deliver another similarly high-profile model capable of rivalling the latest offerings from Google, OpenAI, and Anthropic. In the meantime, competitors such as ChatGPT, Gemini, and Claude have continued to evolve rapidly, while even Chinese players like Alibaba have also gained ground, with its Qwen models performing strongly across various global benchmarks.
However, there have been recent speculations that DeepSeek is set to roll out another major update. The Bloomberg report notes that speculation around a new model launch from DeepSeek has sent ripples through the Chinese tech sector, forcing its rivals like Alibaba, ByteDance, and Tencent to release a barrage of new AI models and services over the Lunar New Year holiday to stay competitive.
DeepSeek had also posted a number of job openings earlier this month which suggested that the company’s latest offering could be in the agentic AI segment. As per another Bloomberg report, the company was looking for specialists in 17 roles, including Agent Deep Learning Algorithm Researcher, Agent Data Evaluation Expert, and Agent Infrastructure Engineer.
Other similar openings required the candidates to become “deeply involved in the application of DeepSeek models in search, content creation, multi-modal, and agent scenarios.”
