How AI Power is Shifting Globally

Source: The post How AI Power is Shifting Globally has been created, based on the article “On AI, the big leap India needs” published in “Indian Express” on 1st March 2025.

UPSC Syllabus Topic: GS Paper3- Science and Technology- developments and their applications and effects in everyday life.

Context: The article discusses the shifting AI power balance, China’s rapid AI advancements, and global competition. It highlights India’s need to lead in AI with top talent, innovation, and a human-centric approach to achieve its goal of becoming developed by 2047.

For detailed information on DeepSeek’s AI Disruption and Lessons for India read this article here

How Did DeepSeek Challenge the AI Landscape?

1. Launch of DeepSeek-R1: DeepSeek, a Chinese startup, introduced its AI model DeepSeek-R1 in January, directly competing with OpenAI’s GPT-4.
2. Open Source and Free: Unlike many other high-performing AI models, DeepSeek-R1 is open source and available for free, making it accessible to a broader audience.
3. Technological Innovation: This development is not just a technological achievement but also a strategic move, showcasing that innovation can be driven by efficiency and strategic planning, not just resources.
4. Shift in AI Power: The emergence of DeepSeek-R1 is a warning to established AI powers, highlighting a shift towards more distributed AI leadership globally.

What Were the Key Outcomes of the AI Action Summit in Paris?

1. Leaders from nearly 100 nations attended the summit to shape AI’s future. They agreed to prioritize AI growth over heavy regulations.
2. France announced a €109 billion AI investment plan to boost competitiveness.
3. President Macron and Finland’s Minister Timo Harakka committed to reducing AI regulations in Europe.
4. India promoted a human-centric AI approach, focusing on jobs, environment, and social well-being.
5. The summit emphasized inclusive and sustainable AI to drive economic growth responsibly.

How is China Advancing in AI Despite Restrictions?

China improved AI by focusing on efficiency due to export controls. Key developments include:

• Pure Reinforcement Learning: AI learns from real-world feedback.
• Mixture-of-Experts (MoE) Architecture: Reduces cost by using fewer parameters per task.
• Multi-Head Latent Attention: Enhances data processing by understanding complex relationships.
  • China also shifted from NVIDIA’s CUDA to PTX to reduce dependence on external technology.

What Steps Did the US Take to Maintain AI Leadership?

1. On January 13, former President Biden introduced an Interim Final Rule to control GPU exports, limiting access to advanced AI technologies for adversaries.
2. On January 14, he signed an executive order to promote clean energy for AI data centers by directing federal agencies to lease land for infrastructure projects powered by low- or zero-carbon energy sources.
3. These measures aimed to secure the US’s AI leadership by controlling critical resources and ensuring AI development is sustainable and resilient.

What is India’s AI Goal for 2047?

1. India aims to become a developed nation (Viksit Bharat) by 2047.
2. The focus is on world-class AI talent to boost global competitiveness.
3. Advanced R&D and innovation are needed to create new technologies, not just use them.
4. At the AI Action Summit, Prime Minister Modi emphasized that AI must serve society responsibly.
5. India’s AI strategy is bold, visionary, and ambitious to lead AI disruption, not just follow.

Question for practice:

Examine how the launch of DeepSeek-R1 has challenged the existing AI power balance globally.