AI: Destiny or tool for shared prosperity

Artificial Intelligence (AI) has moved from the periphery of research labs into the mainstream of everyday life, shaping the apps we use, financial services we consume and the global business models redefining competition. Beneath AI’s growing optimism of efficiency and innovation, however, lies a defining question: Will it reduce global inequalities by broadening overall prosperity, or

will it further entrench wealth divides by concentrating power in fewer hands?

This is not a merely theoretical debate among economists. From policy-makers, central banks, civil society to corporate boardrooms, politicians and leaders are confronting the socio — economic stakes of AI. The evidence so far seems mixed. The outcome will depend less on the growth, acceptance and maturity of technology and more on the governance, incentives and policies that frame its adoption.

Productivity puzzleDespite the transformative potential, AI’s near-term impact on aggregate growth appears modest. MIT economist Daron Acemoglu estimates that baseline productivity gains from its adoption may be as low as 0.5 per cent annually over the next decade. In its 2024 working paper, ‘The impact of artificial intelligence on productivity, distribution and growth,’ the OECD resorts to discordant notes. It feels that AI’s economic contributions will be uneven, benefiting high — tech sectors disproportionately while leaving others behind.

Where disruption is more immediate is in the global labour markets. According to the International Monetary Fund, nearly 40 per cent of global employment will be exposed to AI — related changes, with advanced economies facing greater risks of job substitution compared to the emerging markets. The decisive variable is whether technology functions as augmentation (empowering workers with new skills and capabilities) or replacement (displacing them altogether).

Four pathways Research by global think tanks and multilateral bodies highlights four reinforcing mechanisms through which AI can exacerbate inequality across the world. These include:

Capital capture: AI primarily rewards those who own the models, data and compute infrastructure. If profits stay concentrated in a handful of firms or geographies, income, wealth and productivity gaps will widen among the nations.

Task polarisation: Studies by several American agencies like National Bureau of Economic Research (NBER) and Brookings indicate that AI disproportionately augments high — skill and cognitive work, while automating routine or clerical roles. This widens the income gaps between top and mid-to — low wage earners. This is currently an issue of concern across nations and companies.

Access asymmetry: AI models are predominantly trained in the English language and optimised for the developed — market contexts, which do and can create systemic disadvantages for the emerging economies, smaller firms and non-English — speaking users.

Network concentration: The technology benefits from data — driven feedback loops. Better and more efficient models attract more users, who generate more data, which further improves the models. This dynamic fosters ‘winner-takes-most’ markets. In such a scenario, as is evident from past tech disruptions, monopolies, duopolies and

oligopolies distort the competitive marketplaces. Together, these four forces include risks that can amplify the growing divide between the ‘Have-AIs,’ and ‘Have-Not-AIs.’ 

Inclusive potential However, there are few encouraging signs. The early case studies show that AI can deliver measurable inclusion when applied thoughtfully, carefully and in a calibrated manner. A NBER working paper (2023), ‘Generative AI at work,’ concludes that AI assistants boosted customer service productivity by 14 per cent, with the largest gains among the less — experienced workers.

In banking, AI-powered fraud detection and risk analytics are expanding access to secure financial services for underbanked populations. Yet, at present, these remain pilot projects rather than indicative of system — wide shifts. Scaling inclusive AI requires rethinking incentives, from maximising short — term margins to enabling long — term, broad — based prosperity.

Leadership test AI is forcing newer forms of leadership questions. Will nations harness the technology to broaden shared prosperity? Whether unbridled and non-competitive market forces will concentrate its rewards in the hands of a few individuals, firms or nations? Within the corporate boardrooms, these imply grappling with three uncomfortable and critical trade-offs. These include:

l Short — term profits versus long-term market health. lProprietary advantage versus social legitimacy.lSpeed of automation versus quality of workforce transition.

Investors, regulators and consumers are increasingly scrutinising not just the growth metrics driven by AI, but workforces, as also the distributional and societal impact. The challenge is not just technological. It is not about the best AI tools, apps and models that emerge. Ultimately, it is about governance and leadership.

Indian context For India, the stakes are high. The country has a large workforce in repetitive service roles. It will be extremely vulnerable to tech and AI disruption. But, India has unique strengths too, which include digital infrastructure (Aadhar, UPI), vibrant fintech adoption, and growing AI capabilities and can turn the current risks into opportunities.

For example, in banking, AI can drive financial inclusion. These may be achieved through multiple routes, which include: lEnabling hyper — personalised financial services in multiple regional languages.

lExtending credit access to underserved micro — markets.lEnhancing fraud detection and security across digital channels.lAutomating back — office operations, while empowering the employees for higher-value tasks.Personally, at i — exceed technology solutions, I have witnessed how AI platforms can enable banks to redesign customer engagement, not just for efficiencies, but for accessibility, inclusion and relevance at scale. This process is scalable and can grow into a nationwide and global trend. India can emerge at the forefront of these changes.

Beyond technology If AI is to reduce, rather than increase, wealth disparity, distribution needs to become a design principle. This implies: lEmbedding inclusion in product roadmaps through, for example, AI models that run on low — end devices, and support local languages. lPrioritising augmentation over replacement and use AI to empower human workers and not eliminate them.

lAligning incentives so that one rewards companies not just for their efficiencies but broad — based impact.

lStrengthening governance by treating workforce transition and customer access as key performance indicators.Defining choice AI is not a destiny. It is merely a set of tools whose impact depends on how we deploy them. Leaders, whether in Government, business or finance, must treat distribution not as a moral afterthought but as a measurable business metric.

Handled wisely, AI can democratise access to services, empower workers and unlock new prosperity. Mishandled, it risks entrenching the existing divides and disparities that may last for generations.  The real question is not whether AI will change the world. It already has begun to do so in unimaginable and imagined ways. The more important issue is: Will the society, nations and policy-makers let the technology to further divide us, or use it as a force to inculcate shared prosperity?

Author is a co-founder and CEO, i-exceed.With over 30 years of experience in banking technology, the author is at the forefront of driving global digital transformation. i-exceed is behind Appzillon, a leading digital banking platform used by more than 125 banks in 25 countries