Technology shares collapsed. Giant companies like Meta and Nvidia faced a number of questions about their future. And technology leaders took social media to declare their fear.
And everything was due to a little known beginning of Chinese artificial intelligence called Deepseek.
Deepseek caused waves all over the world on Monday as one of her achievements – that she had created a very powerful model he with much less money than many experts thought – raised a host of questions, including if American companies were Even competitive in him.
Deepseek is the “Sputnik Moment I”, Marc Andreessen, a technology capitalist, posted on social media on Sunday.
How can a company that few people have heard had such an effect?
What is Deepseek?
Deepseek is a start-up and owned by the Chinese High-Fly Stock Trading Firm. Its goal is to build technology of it along the OpenAi chatbot chatbot lines or Google twins. By 2021, Deepseek had won thousands of computer chips by American Nvidia Chipmaker, which are an essential part of any attempt to create powerful systems it
In China, the beginning is known for the abduction of young and talented scholars of him from high universities, promising high salaries and an opportunity to work on front research projects. Both Flyer and Deepseek are led by Liang Wenfeng, a Chinese entrepreneur.
In recent years, Deepseek has released some major language models, which is the type of technology that supports chatbots like chatgpt and twins. On January 10, she released her first free chatbot app, which was based on a new model called Deepseek-V3.
Why did the stock market react to him now?
When Deepseek introduced his Deepseek-V3 model the day after Christmas, it matched the skills of the best chatbots from US companies like Openai and Google. Only this would have been impressive.
But the team behind the new system also revealed a bigger step forward. In a research paper explaining how he built technology, Deepseek said he used only part of the computer chips so that the leading companies were supported to train their systems.
The highest companies in the world usually train their chatbots with super computers using up to 16,000 chips or more. Deepseek’s engineers said they only needed about 2,000 Nvidia chips.
Why is it important?
Since the end of 2022, when Openai began the boom and he, the prevailing notion had been that the most powerful systems could not be built without investing billions of dollars in his specialized chips. This would mean that only the largest technology companies – such as Microsoft, Google and Meta, all of which are based on the United States – can afford to build key technologies.
(New York Times has sued Openai and his partner, Microsoft, claiming copyright violations of news content with respect to he. Both technology companies have denied the claims of the lawsuit.)
But Deepseek’s engineers said they only needed about $ 6 million in raw computing to train their new system. This was approximately 10 times less than what Meta passed the construction of her latest technology.
How did Deepseek her technology with fewer chips he?
He’s top engineers in the United States say Deepseek’s research work put forward smart and impressive ways to build technology with fewer chips.
In short, starting engineers demonstrated a more efficient way to analyze the data using the chips. The main systems of it learn their skills by marking models in large quantities of data, including text, images and sounds. Deepseek described a way to spread this data analysis in some specialized models of him – what researchers call a “mix of experts” method – minimizing lost time by moving data from one place to another.
Others have used similar methods before, but the movement of information between models tends to reduce efficiency. Deepseek did so in a way that allowed him to use less computing power.
“Has it made it very clear that other companies, not just someone like Openai, can build these types of systems,” said Tim Dettmers, a researcher at the Allen Institute for Artificial Intelligence at Seattle and a Professor of Computer Science at Carnegie Mellon University which specializes in the construction of he’s efficient systems. “Deepseek used the method that everyone can copy.”
Deepseek’s research newspaper raised questions whether large US companies can maintain a considerable lead in it many experts believe that technology will become a commodity, with many companies selling the same product.
Is Deepseek technology just as good as systems from Openai and Google?
Deepseek-V3 can answer questions, solve logical problems, and write its computer programs effectively like everything already in the market, according to standard standard tests.
Shortly before Deepseek released its technology, Openai had discovered a new system, called Openai O3, which looked more powerful than the Deepseek-V3. But Openai has not released this system to the wider public.
Openai O3 was created to “reason” through problems involving mathematics, science and computer programming. Many experts pointed out that Deepseek had not built a reasoning model along these lines, which is seen as the future of it
Then on January 20, Deepseek released his own reasoning model called Deepseek R1, and he also impressed experts. This eventually sent us investors and others to a panic at the end of last week and over the weekend, as they understood the importance of Deepseek’s new technology.
American technology giants are building specialized chips data centers. Does this still matter, given what Deepseek did?
Yes, it still matters.
A large number of chips he can still help companies in many ways. With more chips, they can run more experiments while exploring new ways of building it in other words, more chips can still give companies a technical and competitive advantage.
More chips will be needed to operate the new race of that “reasoning”, experts said. These require more computing power when people and businesses use them.
Have the United States have limited the number of Nvidia chips sold in China?
Yes. To maintain the US leadership in his global race, the Biden administration had set rules that limit the number of powerful chips that could be sold in China and other rivals.
But the impressive performance of the Deepseek model raised questions about the unintentional consequences of the US government’s trade restrictions. Checks have forced China researchers to become creators with a wide range of tools that are freely available online.
Some experts continue to argue in favor of the US trade restrictions, saying they were only established recently and that they will have a greater effect on China’s ability to create as the years go by.
Does Deepseek’s technology mean that China is now in front of the United States in it?
No. The world has not yet seen the O3 model of Openai, and its performance in standard standard tests was more impressive than anything else in the market. But experts are concerned that China is being thrown forward in open source systems.
Exactly what exactly is the one with open source?
Like many other companies, Deepseek has a “open source” his latest system of him, which means he has shared the basic computer code with other businesses and researchers. This allows others to build and distribute their products using the same technologies.
This is part of the reason that Deepseek and others in China have been able to build competitive systems so quickly and free.
In the world of open source, steam gathered in 2023 when Meta freely shared a system called Llama. At the time, many assumed that the open-source ecosystem would only flourish if companies like the giant meta-firma with large data centers filled with specialized chips-continued to open the source of their technologies.
But Deepseek and others have shown that this ecosystem can bloom in ways that extend beyond American technology giants.
Why is it important?
Many experts have argued that large US companies should not open their own technologies because they can be used to spread disinformation or cause other serious damage. Some US lawmakers have explored the possibility of preventing or hitting practice.
But other experts have argued that if regulators hinder the advance of open source technology in the United States, China will gain a significant advantage. If the best open source technologies come from China, these experts argue, American researchers and companies will build their systems on those technologies.
In the long run, this can put China in the heart of the research and development of it, which can further accelerate its effort to build a wide range of technologies of it, including autonomous weapons and other military systems.