Published on: March 9, 2025 / update from: March 9, 2025 - Author: Konrad Wolfenstein
The global AI race: Chatgpt too expensive? 700,000 vs. 83,500 euros? 60-hour week for AI victory? Google founder raises the alarm! - Image: Xpert.digital
The global AI race: Innovations and economic tensions in focus
The global AI race: between technological innovation and economic challenges
The technology world is currently in an intensive race for dominance in the field of artificial intelligence (AI). Recent developments show that this competition is led not only between companies, but also between nations - especially the USA and China - with increasing intensity. The surprising progress of Chinese AI models and the reactions of American technology companies illustrate how serious the situation has become and what economic implications this development brings.
Suitable for:
Deepseek's climb: China's surprising advance in the AI
At the end of January 2025, a message shook the global technology landscape: The Chinese start-up Deepseek and its AI assistant R1 won the first place of the most popular free apps in the Apple App Store and thus pushed chatt from Openai from the top position. This development came as a surprise to many experts, since the United States has so far been the undisputed market leader in the field of artificial intelligence.
Deepseek, a AI development laboratory based in Hangzhou, was founded by Liang Wenfeng and specializes in the creation of large language models. What makes Deepseek particularly remarkable is not only the performance of its models, but above all the cost -efficient approach. The company claims to have developed its leading models such as Deepseek V3 and R1 with a budget of less than $ 6 million. This is clear in contrast to the billions of billions that American companies invest in their AI development.
Suitable for:
Rasmus Rothe, co-founder of the Berlin AI investor Merantix Capital, was impressed by Deepseek's performance: "You can tell that the developers were very intelligent." Although the Chinese model has to be subjected to further tests, it already plays “in the same league” as its American competitors.
The question of costs: a crucial competitive factor
A central aspect in the AI race is the economic dimension. The operation of advanced AI models causes enormous costs that influence profitability and market penetration in the long term. It is estimated that Chatgpt's operation costs around $ 700,000 per day, which corresponds to about 36 cents per request. Most of these costs are eliminated by the hardware infrastructure required for the operation of the AI systems.
In contrast, Deepseek chose a remarkably cost -efficient approach. The company states that the total costs for the last training run of your model were only $ 5.576 million. This difference in cost represents a significant competitive advantage and could fundamentally change the dynamics in the AI industry.
What we are watching in the AI race is nothing less than an economic earthquake. The Chinese innovation has added a sensitive defeat to the United States: With operating costs of just 83,500 euros a day, Deepseek undercuts the American flagship chatt for a factor - that is 700,000 euros a day. This cost difference will decide in the long term about market shares and possibly even the technological dominance.
It is particularly noteworthy that Deepseek has achieved this efficiency despite the American export restrictions that should restrict the access of Chinese companies to advanced semiconductor technology. The researchers at Deepseek stated that their Deepseek V3 model was trained with Nvidia's H800 chips, a less advanced alternative that is not affected by the restrictions.
Google's reaction: Brins call to intensify the work culture
The progress of Chinese AI companies has triggered concern in American technology giants. Google, once a pioneer in AI research, was already surprised by the sudden popularity of Chatgpt and was initially unable to bring a product onto the market that could keep up with its awareness.
In this tense situation, Sergey Brin, co -founder of Google, has now caused a stir with an internal memo. According to the New York Times, he asked the employees of the Gemini team-Google's Department of AI models and applications-to increase efforts. "I recommend being in the office every weekday," he said, said the report and described 60 hours a week as the "Sweet Spot of Productivity".
Brin emphasized the urgency of the situation with the words: “The competition has intensified significantly, and the final race for the AGI (Artificial General Intelligence) has started. I think we all have the ingredients to win this race, but we have to accelerate our efforts ”. His statements illustrate how Ernst Google takes the competition and how intensively the race for the management position in AI development has become.
The deepseek effect: vibrations on the financial markets
The effects of deepseek success extend far beyond technological aspects and have also significantly influenced the financial markets. At the end of January 2025, the discussion about the Chinese start-up triggered a real course quake, which became known as a “deepseek effect”.
Suitable for:
The US chip group Nvidia, whose graphics processors have so far been considered indispensable for the training of AI models, was particularly hard. The Nvidia share occasionally fell by 17 percent, which is the greatest loss of value that a company on Wall Street had previously had on a single trading day. Other technology companies, including Siemens Energy and Siltronic, also recorded significant price losses.
These market reactions illustrate how sensitive investors react to changes in the AI landscape and how closely the economic success of many companies is linked to the further development of the AI sector.
USA versus China: The geopolitical context of the AI competition
The race between the USA and China in the field of artificial intelligence is often compared to the race in space at the time of the Cold War. So far, the United States has a lead, especially in the investments. According to a study by the University of Stanford, almost 900 new AI start-ups were founded in the USA last year, compared to 122 in China. At the investments, the United States is far ahead of China with almost $ 70 billion.
Nevertheless, China quickly caught up in certain areas of application. The strength of the Chinese AI developers lies in concrete applications of artificial intelligence and not primarily in Large-length models. Chinese companies show that they catch up technologically in the humanoid robotics.
The American government has tried to slow down China by restricting the export of advanced semiconductor technology. These measures, which were expanded in 2021, aim to prevent Chinese companies from buying powerful chips such as Nvidias A100 and H100, which are often used for the development of large AI models. However, Deepseek's success questions the effectiveness of this policy.
The future of the global AI competition
The race for dominance in the field of artificial intelligence has not yet been decided. While the United States still has structural advantages - not least due to the dominant position of its technology companies and massive financing - the rise of Deepseek shows that innovation and efficiency can challenge traditional strengths.
With Gemini 2.0, Google is preparing for the next phase of this competition. The new model promises improved performance, new skills and a focus on AI agents who can carry out complex tasks independently. With Project Astra, the company also develops a universal AI assistant that will be available on various platforms.
At the same time, Deepseek remains a challenge for the established actors. The ability of the Chinese company to develop competitive AI models with limited resources could change the dynamics of the market sustainably.
Innovation, efficiency and the future of the AI
The current race for leadership position in artificial intelligence is characterized by dynamic changes. Deepseek's surprising success has shown that innovation and cost efficiency can challenge established market leaders. At the same time, Google's reaction illustrates how seriously the American technology companies take this challenge.
The economic dimension of this competition is becoming increasingly important. The enormous costs for the development and operation of advanced AI models are a challenge that can only be mastered by continuous innovation and increase in efficiency. Companies that successfully master this balance will play a leading role in the future of AI.
Ultimately, the users of AI technologies benefit from this competition. The intensified competition drives the development of more powerful, cheaper and more accessible AI solutions that have the potential to transform numerous aspects of our daily life and the global economy.
Suitable for:
Your global marketing and business development partner
☑️ Our business language is English or German
☑️ NEW: Correspondence in your national language!
Konrad Wolfenstein
I would be happy to serve you and my team as a personal advisor.
You can contact me by filling out the contact form or simply call me on +49 89 89 674 804 (Munich) . My email address is: wolfenstein ∂ xpert.digital
I'm looking forward to our joint project.
