DeepSeek, a Chinese AI startup, triggered a broad selloff in global markets after its new reasoning model and chatbot app convinced investors that advanced AI could be built with far less compute and money than many U.S. rivals had assumed. On January 27, tech stocks fell sharply, with Nvidia leading losses and the Nasdaq futures also dropping as traders reassessed the outlook for AI chip demand and the valuations of major AI-linked companies. The central market concern was not just that DeepSeek had released a competitive model, but that it appeared to do so at a fraction of the cost typically associated with frontier AI development. Reports cited DeepSeek’s claims that its model matched leading U.S. systems while using fewer Nvidia chips and a much smaller training budget, which intensified doubts about the scale of future spending on GPUs, cloud infrastructure, and other AI inputs. The episode mattered because it challenged a core assumption underpinning the recent AI boom: that progress in frontier models would continue to require massive capital expenditure by a small set of U.S. firms.

AI-generated background, compiled from web sources — not editorial content.

More coverage

Explore the topic

More on $BTC

Comments