DeepSeek dominates AI industry and causes S&P 500 and Nasdaq to fall
Chinese artificial intelligence (AI) start-up DeepSeek, which only employs 200 people, has resulted in a member of the Magnificent 7 losing $600bn of market value and the US president declaring it a “wake-up call” for US tech firms.
Deepseek has proved two main points that have hurt other US AI firms, according to Edmond de Rothschild AM co-head of equities Jacques-Aurélien Marcireau. Firstly, its cost of computing is far cheaper, with a “shoestring budget of under $6m”.
President Trump recently announced $500bn investment over four years into Stargate, which will make Artificial General Intelligence a reality and lead to truly sentient AI.
Stargate was already being discussed by OpenAI, Microsoft and Oracle for almost a year, prior to the US president’s announcement.
Secondly, Marcireau said it is open source, “meaning that instead of cloaking advances in secrecy, China is actively contributing to the open-source community, sharing knowledge for everyone’s benefit”.
In response to DeepSeek’s meteoric rise, Nvidia, a member of the Magnificent 7, lost $600bn of market value – the biggest ever loss for a stock in a single day, according to Hargreaves Lansdown head of equity research Derren Nathan.
Syz Group CIO Charles-Henry Monchau had predicted that the Mag 7 will turn into the Lag 7. However, DeepSeek has accelerated this process.
Monchau said: “We estimated mid-2025, market sentiment would pick up that we were facing significant competition from China. We felt this development would trigger a sharp decline in Nvidia’s stock price, mirroring the broader downturn affecting the other Magnificent 7.”
DeepSeek has pushed mid-2025 to the start of 2025.
Monchau added: “The new developments have raised alarms on whether America’s global lead in AI is shrinking and called into question big tech’s massive spend on building AI models and datacenters.”
Other big semiconductor names that have been caught in the crossfire include custom chip designer Broadcom and memory specialist Micron. Outside of the US, stocks that have taken a hit range from Taiwan Semiconductor Manufacturing Company through to ASML, the Dutch builder of chip-printing machines.
Liontrust head of the global equities team Mark Hawtin said that “in our view, the DeepSeek buzz is justified and more notable when considering the US imposed restrictions on Nvidia’s AI chips to stall AI advancement in China in an ‘AI Arms Race’”.
Hawtin added this “forced efficiency” led to the creation of “novel architectures such as a Mixture-of-Experts (MoE), which only activates the necessary parameters and optimises communication between GPUs”.
DeepSeek employs techniques such as DualPipe algorithms to reduce training costs. It also uses an approach called multi-head latent attention, which compresses attention mechanisms to minimise memory usage during inference, enhancing speed without sacrificing accuracy.
Importantly, Hawtin explained that DeepSeek’s open architecture is not just cheaper, it is also more powerful.
“We could witness an explosion of new applications by decoupling AI advancement from high capital costs.”
Fidelity International investment director Tom Stevenson said: “Wall Street reeled at today’s (27 January) opening, with the S&P 500 down 2% and the Nasdaq tumbling 3.5%.”
DeepSeek rivals OpenAI and Meta, while relying on significantly fewer Nvidia chips. It was founded in 2023 as an offshoot of a quant hedge fund and has quickly risen to fame with its rapid release of efficient open-source AI models challenging the Western tech giants.
Its DeepSeek-V3 model, released in December 2024, was highly regarded for its cost effectiveness.