I’m sure by now, you’ve all seen the damage a Chinese AI company did to the markets yesterday.

Using only 2000 older GPUs, they produced an AI model with the same performance as the leading US models in a matter of months.

Now, why is this so important?

👉It undermines the position of the leading GPU manufacturer, NVIDIA. They produce the default chip for training and running AI models. One chip easily costs $80k, and the big players bought 100,000. Deepseek didn’t have access to these as the US banned exports to China, but it was still able to build a leading model.

👉It pressures the other model providers to lower their usage prices. For example, the reasoning model from OpenAI is 10 times more expensive to use than Deepseek.

👉It questions the investments that big players have made. Some have spent hundreds of millions of dollars to build a model, while Deepseek did it with $5 million, and they open-sourced their model at the end.

👉It will also affect other stocks/players, such as the energy sector and the need for multiple, huge data centers…

It is important to note that although there are technical and financial implications, this doesn’t mean that Deepseek will now take over. It is still a Chinese company that has to follow the Chinese Security law on data, similar to the controversy around TikTok.

Also, a lot of these claims still have to be verified.
Let’s wait and see what happens while the big players think about their next moves!