The release of DeepSeek, a Chinese-made generative AI chatbot similar to ChatGPT, has been heralded as “AI’s Sputnik moment.”
Since its launch in December 2024, DeepSeek has topped the Android and iOS app store charts.
Completing its training in just two months for a mere $5.5 million, DeepSeek underscores China’s strategic efficiency in AI development, optimising performance while reducing costs and computational demands.
This development reignites memories of the space race and introduces international competitiveness to AI development – a field that, until now, has been dominated by US firms.
DeepSeek’s advancements, particularly with its latest models V3 and R1, have been praised for their superior natural language processing, writing quality, and reasoning capabilities.
These developments showcase how China, despite firms facing restrictions on cutting-edge semiconductor imports, has managed to compete with western AI technologies.
DeepSeek’s V3 was trained on Nvidia H800 GPUs – less advanced and cheaper than the H100 GPUs used by Western AI giants.
The emergence of DeepSeek has raised questions about the substantial investments in American AI firms, especially as the release of these models led to a staggering $1 trillion being erased from the American stock exchange in a single day due to mass sell-offs.
DeepSeek’s ascent is not an anomaly. Since mid-2023, Chinese tech giants like Alibaba, Tencent, and ByteDance have been narrowing the technological gap with their American counterparts.
These companies are not only catching up in terms of technological capabilities but are also surpassing competitors through cost efficiency.
With a robust ecosystem for AI research and development continuing to grow, China is well-positioned to reshape the global AI dynamics in the forthcoming years.
“Machine Learning & Artificial Intelligence” by mikemacmarketing is licensed under CC BY 2.0.