1. Identify a nodal agency within the government to lead the charge on AI
According to study, there needs to be a leading agency that can coordinate all of India’s AI activities. That same agency will also be responsible for pushing other departments within the government to develop their own AI capabilities.
Having a central agency in place will allow the government to prioritise its resources and give support to socially relevant applications.
2. Build a collaborative framework for engagement between governments, industry and academia
Once a nodal agency is put in place, networks have to be established. Departments at state and national levels should be able to talk to each other in order to produce AI-driven solutions which are locally relevant — but also encourage growth.
The study points out that a significant push needs to be given the hardware sector that’s lagging behind.
3. Build an AI data strategy
In other words, to get an AI algorithm to function at its best, it needs data. But data isn’t easy to come by. Even if you have access to some, there are multiple questions around its ethical use and privacy.
Which is why the government needs to be at the helm to ensure that AI-compatible data can be made available to entrepreneurs. In addition, the data that’s already publicly available to enable integration for AI — not just literature.
Address the skill-gap
Case studies show that while some businesses have been able to hire talent easily, others have struggled. The lack of AI courses and updated skill-sets resulted in a skill-gap that hampers the growth of AI in India.
Considering that the applications of AI go beyond technology, its integration into curriculums shouldn’t be restricted to technology-related courses either. NASSCOM’s study suggests creating a marketplace where AI professionals can gather in one place to address the gap that start-ups are currently facing.
Keep it safe and unbiased
It’s not just India that’s struggling with governance around the use of data — it’sa global issue. However, in order to address it, India needs to find new ways to be an active participant in the AI standardization process.
The study also recommends setting up Algorithmic Impact Assessments (AIA) to guard against biased outcomes and their potential impact.