In just a week, the startup has sent shockwaves across the industry after its AI assistant overtook OpenAI’s ChatGPT as the most-downloaded free app in the US, causing stocks to crash.
just launched its R1 model last week and has quickly positioned itself as a competitor to major players like OpenAI and Google in the generative AI space.
It has been lauded for its reasoning capabilities and cost efficiency. Industry experts estimate the model’s development costs to be under US$6 million—less than 10% of the training expenses of comparable models like Meta’s Llama. Built despite US restrictions on chip exports to China, R1’s performance has propelled the company into the spotlight, sparking discussions on its implications for the global AI arms race.
The buzz around DeepSeek has been amplified by its unique approach. Originating as part of a Chinese hedge fund’s AI research unit, the startup transitioned into developing advanced large language models and pursuing artificial general intelligence (AGI). Its open-source R1 model has garnered significant praise from developers and analysts for its accessibility and efficiency.
DeepSeek’s meteoric rise has prompted broader questions about the AI industry’s sustainability. With soaring funding rounds and valuations dominating the sector, the startup’s rapid progress at relatively low cost challenges assumptions about the necessity of billion-dollar investments. As the generative AI market hurtles toward an estimated US$1 trillion in revenue within the next decade, DeepSeek’s trajectory invites questions on the industry’s growth.
