"What Deepseek showed the world, beyond causing a big hiccup in the financial market valuation of the US AI tech giants, is that China, which the US had attempted to undermine by only allowing the export of less powerful AI acceleration hardware, was able to produce a model that could outperform US LLMs that benefited from the most powerful Nvidia chips."
"Its significance goes beyond geopolitics: Deepseek's R1 model demonstrated that it is certainly not necessary to throw vast amounts of computational resources and spend a huge amount of money on AI acceleration hardware to achieve good results. The financial results from the hyperscalers show that the trend is to invest heavily in gigawatt datacentres, which they anticipate will be needed to support the most powerful AI acceleration hardware."
Deepseek's R1 model from China outperformed US large language models despite US restrictions on exporting powerful AI acceleration hardware. The model challenged assumptions that massive compute and expensive accelerator hardware are necessary for top results. Hyperscalers are investing in gigawatt datacentres to host powerful AI accelerators, but such infrastructure is excessive for many organizations. Smaller, focused models can combine public LLM capabilities with targeted training to deliver superior business outcomes. Agentic AI became a dominant 2025 trend, prompting vendors to embed AI in enterprise software. Many corporate AI projects show poor ROI and fail beyond pilots, while standalone AI features proliferate across IT systems.
Read at ComputerWeekly.com
Unable to calculate read time
Collection
[
|
...
]