India’s AI Industry: Balancing Power, Openness, and Human Augmentation
The world’s big artificial intelligence (AI) companies are raising concerns and calling for regulation, similar to how incumbent telcos used to promote ‘Fear, Uncertainty, and Doubt’ (FUD) 20 years ago. It is reasonable to question whether their motivations are driven by the desire to secure their dominant market positions rather than a genuine concern for public safety. OpenAI and Google’s calls for regulation of foundation models could be an attempt to lock in their positions and hinder new entrants.
The narrative surrounding the dangers and risks of generative AI seems to be influenced by a few big companies in Silicon Valley and the intellectual and media ecosystems they have fostered. While the concerns about the future impact of artificial general intelligence and the work being done to ensure alignment with human interests are important, they can divert attention from more immediate questions of public policy.
Like other countries, India faces the challenge of governing the AI industry, which is already here, complex, and not well-defined. Delaying the implementation of a governance framework allows first movers to establish their positions and practices, potentially limiting the government’s ability to shape the industry. In this regard, it is important to focus on governance rather than regulation to allow for a flexible approach to policy-making.
India should adopt several principles to govern the AI industry. Firstly, there is a need to ensure that power is not concentrated in the hands of a few. By making AI accessible to everyone, it prevents the technology from being used to dominate others. While there will be individuals who exploit the power for malicious purposes, the positive impact of widespread access to AI will outweigh the negative.
Secondly, India should prioritize open technologies over proprietary ones. Open-source software, models, data-sets, and digital standards can make technology more affordable and lower barriers for Indian technology companies to compete globally. Focusing on open-source projects and building for the whole planet can give India a strategic advantage in the evolving technology landscape.
Thirdly, as a labor-surplus economy, India should favor AI systems that augment human capacity rather than replace it. Augmenting labor creates new value and can lead to increased employment, productivity, and real wages. Given the potential impact of AI on jobs, finding the right balance between automation and augmentation is crucial for India.
To translate these principles into a governance framework, the AI industry can be divided into four layers: data, models, computation, and application. It is important to ensure competition within each layer and prevent vertical integration across layers. This can be achieved by requiring data owners, model developers, and computation providers to make their resources available on the same terms to promote fair competition.
India has a moral imperative to seize every opportunity to improve the outcomes for its people. While concerns about the harms of generative AI exist, it is crucial not to succumb to fear, uncertainty, and doubt. Instead, policymakers, experts, and technologists should focus on creating an environment that fosters innovation, fairness, and widespread access to AI.
In conclusion, India must navigate the governance of its AI industry by ensuring power is not concentrated, favoring openness over proprietary technologies, and promoting human augmentation. By establishing a governance framework that encourages competition within each layer of the AI industry, India can shape the future of AI to benefit its citizens. Embracing AI while addressing potential risks and challenges will be vital for India’s growth and development in the digital age.
