The recent impact of DeepSeek’s advanced AI model, which triggered a significant drop in US tech stock values, has spurred India into action. IT Minister Ashwini Vaishnaw’s announcement of substantial compute subsidies, increased startup funding, and other supportive measures underscores the government’s commitment to accelerating India’s AI capabilities and bridging the gap with global AI leaders. However, government support alone is insufficient. India’s tech sector needs a fundamental shift in mindset, drawing inspiration from China’s approach, to truly compete in the rapidly evolving AI landscape.
A telling anecdote highlights the challenge. During OpenAI CEO Sam Altman’s visit to India in 2023, he was asked about the feasibility of a small team with limited funding ($10 million) developing a foundational AI model comparable to ChatGPT. Altman’s response was twofold. Initially, he dismissed the idea as “hopeless,” suggesting the immense resources and expertise required for such an undertaking. However, he added a crucial caveat: “It’s your job to try anyway.”
Unfortunately, the Indian tech community seems to have fixated on the first part of Altman’s statement. Discussions with venture capitalists and founders over the past year reveal a prevailing focus on leveraging existing foundational models developed by large American AI companies. These models, backed by colossal investments, are seen as the primary building blocks for Indian AI initiatives. While this approach can yield valuable applications and advancements, it risks limiting India’s potential to become a true innovator in the field.
Emulating China’s approach requires more than just financial investment. China’s success in AI stems from a combination of factors, including a strong emphasis on research and development, a supportive regulatory environment, and a culture that encourages ambitious, long-term projects. Indian tech needs to cultivate a similar spirit of innovation and a willingness to take on challenging, foundational problems.
Several key areas require attention:
- Investing in Fundamental Research: While building on existing models is important, India needs to significantly increase investment in fundamental AI research at universities and research institutions. This will create a pipeline of skilled researchers and foster the development of indigenous foundational models.
- Nurturing Deep Tech Startups: Startups focusing on core AI technologies, including algorithm development, hardware acceleration, and data infrastructure, need targeted support. This requires more than just funding; it also involves providing mentorship, access to resources, and a regulatory environment that encourages innovation.
- Promoting Collaboration: Collaboration between academia, industry, and government is crucial. This will ensure that research efforts are aligned with industry needs and that innovations can be quickly translated into practical applications.
- Developing Talent: India has a large pool of talented engineers, but specialized AI talent is still in high demand. Investing in AI education and training programs is essential to build a robust AI workforce.
- Embracing a Long-Term Vision: Developing cutting-edge AI capabilities is a marathon, not a sprint. India needs to adopt a long-term vision and commit to sustained investment and support for AI research and development.
While leveraging existing models can provide a starting point, India’s true potential lies in developing its own foundational models and pushing the boundaries of AI research. This requires a shift in mindset, embracing ambitious goals and fostering a culture of innovation. Only then can India truly become a global leader in the AI revolution.