Editorial

Artificial intelligence (AI) is the latest stop in our journey of jumping from one technological paradigm to another—from the World Wide Web to mobile phones and smartphones—each hailed as a revolution that would empower societies. However, these revolutions have also left many people behind or further marginalised, reinforcing existing inequalities even as they created new opportunities. Unless designed and governed inclusively, AI risks repeating this pattern—another powerful innovation that will likely benefit the privileged while leaving the poor and marginalised dependent on new systems.

Volume 11 Issue 1 of the Journal of Development Policy and Practice explores the theme of ‘Algorithms, AI & Accountability’, with contributions that examine the promises, perils and ethical implications of AI. Collectively, they pose a crucial question: Can AI promote democracy and development, or will it exacerbate the crisis of exclusion? These debates are not merely technical but profoundly ethical: Who is accountable when AI makes a decision, and how do we ensure consent and dignity in a world where people risk being treated as mere data points?

We begin with two commentaries. Hemant Adarkar’s Preparing India for AI Adoption identifies both opportunities and systemic challenges in adopting AI at scale, while Siddhartha Malempati’s The Body Is Not a Password offers a powerful critique of biometric authentication. Malempati warns of a world where people themselves are reduced to data points, stripped of complexity and dignity—a fear that runs through this issue.

The research papers that follow examine these issues across different fields. Mahera Imam investigates Gender Dynamics and the Misuse of Digital Technologies, demonstrating how women often experience technology as an added layer of control. Medhavi Gupta’s Allusions of Dystopian Techno-optimism questions the cultural ideas surrounding AI assistants like Siri and Alexa, revealing how convenience stories hide corporate reliance. Christopher High considers AI in agriculture, balancing its potential benefits for small farmers against the risks of new dependencies. Dharish David, in Algorithmic Bias and Discrimination, shows that bias is not an anomaly but a fundamental aspect of AI systems. Nahia Hussain’s Digital Health Inequities points to the unequal adoption of digital health tools during the pandemic, serving as a warning for future AI implementation.

The issue concludes with Sejal Gupta’s review, India’s AI Regulation at the Crossroads, which places India’s policy options within the global regulatory landscape. It highlights that regulation must be proactive, inclusive and mindful of the country’s complex social demography if AI is to support, rather than undermine, democracy.

Across these contributions, a broader picture emerges. Platformisation has already failed to provide adequate answers to the needs of equity and democracy; layering AI upon it risks the dehumanisation of society, where people are increasingly managed as data and automated actors. AI is essentially an extractive technology, drawing data relentlessly from people, citizens and communities while offering little in return. The irony is that rapid digitalisation has deepened the digital divide, and the emerging AI divide is built on this digital divide, which in turn is rooted in long-standing social divides tied to unequal access to resources and infrastructures.

In India, where caste, class, gender and language create deeply stratified realities, poorly designed AI systems will amplify existing exclusions. And as AI development is driven largely by global corporations and technology giants, there is a real danger that democratic oversight will be eroded, leaving development agendas subordinated to corporate priorities.

For AI to rise above another hype cycle, its ethical foundations must be bolstered. This involves building in safeguards during design, broadening data sets, and establishing regulatory frameworks that safeguard the interests of marginalised communities. It also means acknowledging that technology is never impartial and ensuring that AI serves people, rather than the other way around. A one-size-fits-all model cannot work in India, a society marked by ladders of social hierarchies and specific lacks and needs. A hyperlocal and decentralised approach—as demonstrated in the SoochnaPreneur ¹ model—must be nurtured alongside co-building digital trust and consciousness. Communities must see themselves as rightful claimants of accountability from institutions and platforms shaping the AI ecosystem. Only then can inclusive AI design be more than rhetoric.

As such, this issue serves as both a warning and a rallying cry: A warning against unthinking enthusiasm that overlooks the risks of AI, and a call to reconsider AI as a tool for promoting fairness, inclusion and democracy. Safeguarding the future requires not only regulatory will but also the active engagement of civil society, researchers and citizens to insist that AI remains a tool for human development rather than human replacement.