China’s DeepSeek AI Moves The Capital Of Tech From Palo Alto To Hangzhou..

In a matter of days, the news of China’s AI sensation, DeepSeek R1, has gone from a gentle breeze to a Force 5 hurricane. It’s clear now that no one in Silicon Valley or Washington DC had the slightest idea that their world was about to be turned upside-down by an innovative new product that would shift the geopolitical plates further eastward..

But that, in fact, is what has happened. And it’s not simply because DeepSeek’s latest version matches or exceeds the performance of America’s best model, OpenAI; but because it is cheaper, more accessible and more transparent. This is AI for everyone regardless of their station or income. And its sudden emergence from ‘out of the blue’ has cast doubts on the ability of western tech giants to anticipate the capability of their competitors or to lead an industry that is essential for Washington to preserve its ever-loosening grip on global power. Here’s a brief recap from Venture Beat:

….thanks to the release of DeepSeek R1, a new large language model that performs “reasoning” similar to OpenAI’s current best-available model o1 — taking multiple seconds or minutes to answer hard questions and solve complex problems as it reflects on its own analysis in a step-by-step, or “chain of thought” fashion.

Not only that, but DeepSeek R1 scored as high or higher than OpenAI’s o1 on a variety of third-party benchmarks…, and was reportedly trained at a fraction of the cost…, with far fewer graphics processing units (GPU) under a strict embargo imposed by the U.S., OpenAI’s home turf.

But unlike o1, which is available only to paying ChatGPT subscribers of the Plus tier ($20 per month) and more expensive tiers (such as Pro at $200 per month), DeepSeek R1 was released as a fully open source model, which also explains why it has quickly rocketed up the charts of AI code sharing community Hugging Face’s most downloaded and active models. 

– Why everyone in AI is freaking out about DeepSeek, Venture Beat

Did you notice that “DeepSeek R1 scored as high or higher than OpenAI’s o1 (while) under a strict embargo imposed by the US”?

In other words, these Chinese whiz-kids created their cutting-edge version with one hand tied behind their back. They shrugged off Washington’s onerous sanctions and beat Uncle Sam at his own game, which is quite an accomplishment. (Forbes: “U.S. export controls on advanced semiconductors were intended to slow China’s AI progress, but they may have inadvertently spurred innovation.”) Here’s more:

thanks to the fact that it is fully open source, people have already fine-tuned and trained many multiple variations of the model for different task-specific purposes such as making it small enough to run on a mobile device or combining it with other open-source models. Even if you want to use it for development purposes, DeepSeek’s API costs are more than 90% cheaper than the equivalent o1 model from OpenAI.

Cheaper, more adaptable and more transparent. Is there more? There is:

Most impressively of all, you don’t even need to be a software engineer to use it: DeepSeek has a free website and mobile app even for U.S. users with an R1-powered chatbot interface very similar to OpenAI’s ChatGPT. Except, once again, DeepSeek undercut or “mogged” OpenAI by connecting this powerful reasoning model to web search — something OpenAI hasn’t yet done…