Syllabus
GS Paper 3 – Awareness in the fields of IT, Space, Computers, robotics, nano-technology, bio-technologyand issues relating to intellectual property rights.
Context
Not aiming to compete with or replicate the Big Tech model is essential; it should even look at championing ‘small AI’.
Source
The Hindu| Editorial dated 23rd November 2024
Democratising AI needs a radically different approach
Big Tech companies increasingly dominate the global Artificial Intelligence (AI) ecosystem. While countries like India are investing in sovereign cloud infrastructure, open data platforms, and local start-ups, such initiatives may inadvertently reinforce Big Tech’s influence unless a fundamentally different approach is adopted.
Challenges of Big Tech Dominance
High Computational Costs
- Deep learning models like Gemini Ultra cost $200 million to train (2023), making it unaffordable for smaller players.
- Entrants often rely on Big Tech’s compute credits, reinforcing their dominance.
- Big Tech advocates for larger models to maintain their market position and recover costs through services.
Superior Infrastructure and Tools
- Big Tech provides a comprehensive end-to-end service offering with optimised workflows, tools for data preparation, and cutting-edge algorithmic models.
- Switching to other providers incurs higher costs, discouraging alternatives to Big Tech infrastructure.
Data Monopolies
- Big Tech enjoys a constant stream of diverse and global data, enabling advanced data intelligence that smaller players cannot match.
- Public data initiatives, though promising, are vulnerable to commercial capture by resource-rich Big Tech companies.
Declining Role of Academia
- Big Tech now leads in academic research and citations, driving the direction of AI development.
- This diminishes academia’s ability to challenge or innovate independently from industry priorities.
The Need for a Paradigm Shift
Re-imagining AI Development
- AI development should focus on purpose-driven, smaller models guided by domain expertise and lived experiences rather than relying solely on statistical patterns in Big Data.
- This would promote democratic and targeted AI development that addresses specific societal challenges.
Theory of Change
- A theory-driven approach focuses on understanding causal mechanisms and testing hypotheses to achieve meaningful change.
- Targeted data collection can refine these models and reduce reliance on large-scale Big Data.
Lessons from Other Fields
- Historical advancements in medicine, aviation, and weather forecasting relied on theory-driven methods, prioritising scientific rigour over sheer data volume.
- A return to this approach could yield innovative and context-specific AI solutions.
Risks of Staying the Current Course
Dependence on Big Tech
- Continuing to prioritise Big Data and deep learning will deepen reliance on Big Tech companies.
- The Global Development Compact, despite its goals of democratising AI, perpetuates the belief that large datasets and computational access are the only solutions.
Wasted Public Resources
- Public investments in replicating Big Tech’s model risk diverting resources from more innovative and locally relevant AI projects.
- A focus on scaling down AI development could prevent this misuse.
Conclusion
Breaking Big Tech’s hold over AI requires a fundamental shift in approach. A model rooted in theory of change and small AI can democratise AI development while promoting innovation and sustainability. Rather than competing with Big Tech on its terms, countries must redefine the rules of the game, focusing on inclusive, purpose-driven AI solutions that prioritise societal impact over sheer scale.
Related PYQ
Introduce the concept of Artificial Intelligence (AI). How does AI help clinical diagnosis? Do you perceive any threat to privacy of the individual in the use of Al in healthcare? [ UPSC Civil Services Exam – Mains 2023]
Practice Question
Discuss the dominance of Big Tech in the Artificial Intelligence ecosystem and suggest strategies to democratize AI development in India. [250 words]
Guidelines for Answering the Question
- Introduction:
- Briefly define Big Tech dominance and its implications for AI development.
- Highlight the significance of democratizing AI for a balanced technological and socio-economic ecosystem.
- Body:
- Discuss challenges of Big Tech Dominance
- Suggest Strategies for Democratizing AI
- Stress the importance of inclusive and sustainable AI innovation for India’s technological sovereignty.
- Conclusion:
- Summarize the need for redefining AI development to reduce dependence on Big Tech.