China’s AI Crown Jewel Just Went Dark for 7 Hours and Nobody Knows Why
2 min readIf you tried to open DeepSeek this morning and got nothing, you weren’t alone. China’s breakout AI chatbot — the one that rattled Silicon Valley and sent Nvidia’s stock into a tailspin last year — just experienced its longest outage since launch: seven hours and thirteen minutes of complete silence, affecting an estimated 355 million users worldwide.The incident began Sunday evening, with users flooding Downdetector to report errors. DeepSeek’s status page acknowledged an initial issue at 9:35 p.m. Beijing time, briefly marked it as resolved, then had to return to address a second wave of performance issues that wasn’t fully cleared until 10:33 a.m. Monday morning. Seven hours of the world’s most-hyped AI platform — gone.
What This Really Means
DeepSeek built its reputation on doing more with less — training world-class models at a fraction of what U.S. labs were spending. That scrappy efficiency narrative was compelling. But “efficient” and “reliable” are two very different things, and right now, DeepSeek has a gap between them.As AI moves from novelty to infrastructure, uptime is everything. Enterprises, developers, and consumers don’t just want a model that can outperform GPT-4 on a benchmark — they need one that’s there when they need it. A seven-hour blackout with no public explanation isn’t just an inconvenience; it’s a trust problem.The cause of the outage remains unknown. DeepSeek has not publicly disclosed what went wrong, which only amplifies the concern. For a platform carrying the weight of 355 million users — and the reputation of China’s entire AI ecosystem — transparency in a crisis matters as much as raw technical performance.The AI race isn’t just about who builds the smartest model anymore. Increasingly, it’s about who can keep the lights on.
Source: Bloomberg, Business Standard