DeepSeek bursts the AI bubble, ushering America into a tech ice age?
In the history of technological development, the world has witnessed the collapse of the Dotcom bubble in the early 2000s. At that time, billionaire Jeff Bezos, founder of Amazon, declared, “Your venture capital is my opportunity.” Today, a similar story seems to be unfolding with the emergence of DeepSeek, an AI model from China.
Affordable AI Tool: The Breakthrough from DeepSeek
Developed by Liang Wenfeng, DeepSeek has shaken the market with the claim, “Your cost structure is my opportunity.” This assertion is not arbitrary but is based on groundbreaking advancements in AI approach and development compared to U.S. tech giants.
Strategic Differences in AI Development
Companies like OpenAI, Google, and Microsoft have relied on AI models since 2017, which have consumed vast computational resources despite continuous improvements. Last year, the top four AI investors in the U.S. spent $180 billion on developing these models, a 57% increase from 2023. This year, Microsoft is expected to invest an additional $80 billion, and Facebook about $65 billion.
However, this strategy has led to the formation of an AI bubble with hundreds of billions of dollars invested but with disappointing results. Many GPUs used to run AI models, along with continuously operating data centers, have not met their initial goals.
DeepSeek: An Efficient Low-Cost Solution
On January 20, DeepSeek officially launched, bringing a fresh breeze to the AI industry. The tool not only performs well against ChatGPT, Gemini, or Llama but also completes tasks much faster. Notably, DeepSeek does not require large GPUs, massive data centers, or abundant power.
With costs up to 20 to 100 times lower than current solutions, DeepSeek promises to change the global AI landscape.
DeepSeek: Paving the Way for Comprehensive Innovation
Open Source Feature
One of the standout advantages of DeepSeek is its open-source feature. This allows all developing countries to use and build independent AI applications without relying on U.S. infrastructure. As a result, countries can develop AI models tailored to their specific needs and conditions.
Resource Efficiency and Anti-Monopoly
DeepSeek not only conserves resources but also contributes to anti-monopoly efforts. The hardware, infrastructure, and energy required to operate DeepSeek enable countries to build sovereign AI models while reducing costs. This facilitates adjustments by businesses, research organizations, and governments to meet specific needs.
Decentralization
Unlike U.S.-based AI models that rely on centralized infrastructure controlled by a few tech giants, DeepSeek is decentralized. This not only enhances flexibility but also reduces dependency risks on a limited number of providers.
Conclusion
The emergence of DeepSeek marks a significant turning point in the AI industry. With its open-source model and superior resource efficiency, the tool challenges the dominance of U.S. tech giants and opens opportunities for other countries to participate in the technology race.
If previously Nvidia was nearly the sole company capable of producing powerful processors for AI, now China has demonstrated that it can achieve the same. Many experts believe that with the advent of DeepSeek, a tech ice age is beginning, and the U.S. may lose the economic competitive edge it staked over the past two years.
For more information about DeepSeek and the latest tech trends, follow COCC-EDU-VN regularly.
Source: dantri.com.vn
