Kosambi’s insights offer us a critical lens to examine the trajectory of artificial intelligence (AI) and its potential to liberate or oppress. On July 31, 2025, we commemorate the 118th birth anniversary of D D Kosambi whose insights on the interplay of science, freedom and society remain strikingly relevant. In his 1949 essay on ‘Imperialism and Peace, Science and Freedom’, which was written against the backdrop of a world that was still coming to terms with inhuman consequences of technology-driven devastation, Kosambi articulated a profound connection between science and societal structures. He argued that scientific progress is not merely a pursuit of knowledge but a reflection of the societal necessities and constraints of its time.
Today we stand at the cusp of an AI-driven revolution and Kosambi’s insights offer us a critical lens to examine the trajectory of artificial intelligence and its potential to liberate or oppress and the urgent need to align it with the broader good of humanity.
Kosambi defined freedom as the “recognition of necessity” and science as the “cognition of necessity.” For him science was not an abstract endeavour but a materialistic, testable process that responds to societal needs. He illustrated this with examples like the invention of airplanes that required understanding aerodynamics and propulsion, and eyeglasses that demanded advances in optics and manufacturing. However, he emphasised that access to these scientific solutions often hinges on economic power, revealing how social structures shape the fruits of science.
In the context of artificial intelligence, this perspective is illuminating. Artificial intelligence, like the flying machines or corrective lenses in Kosambi’s examples, represents a monumental leap in humanity’s ability to address complex problems, from healthcare diagnostics to climate modelling. Yet, just as millions in 1949 India lacked access to glasses due to poverty, millions today continue to be excluded from AI’s benefits due to economic, digital or educational divides.
In 1949, Kosambi observed that American scientists, despite their concerns about “scientific freedom,” were increasingly beholden to big business, war departments or universities funded by them. He argued that science had lost its independence, becoming a tool for profit-driven patents or militaristic agendas, with scientists facing pressures to avoid “dangerous” ideas such as those labelled communist during the McCarthy era.
Today, AI development mirrors this dynamic. Major AI systems, including large language models and generative tools, are predominantly controlled by a handful of tech giants and governments. These companies prioritise commercial applications like advertising algorithms, surveillance systems etc over societal needs like equitable education. The modern AI scientists also operate within a tightly integrated system where funding and priorities are determined by those who “pay the piper” as Kosambi said. For instance, with companies like those behind ChatGPT, DeepMind or Grok focus on models that generate revenue rather than open-source solutions that could democratise access.
Kosambi’s call for science to serve all of humanity and not just the elite is a clarion call for AI’s future. He wanted science which was liberated from servitude to profit and warfare. Artificial intelligence has the potential to continue on the trajectory of automating tedious tasks, accelerating medical research. However, Kosambi would caution against its misuse.
The development of AI for mass surveillance, psychological manipulation and autonomous weapons echoes his warning about scientists serving a class that “dumps food in the ocean while millions starve”. Experts at Internet Governance Forum 2025 Open Forum highlighted urgent need to counter AI-driven disinformation threats to democracy. These disinformation campaigns are likely to affect disproportionately marginalised communities and vulnerable groups. Hence, there is a need for an immediate course correction.
Looking at AI through Kosambi’s framework, it needs to be ensured that AI’s benefits are not just confined to the digital elite. In the case of India, the “Digital India” could prioritise AI infrastructure that could serve rural and underserved communities such as AI aiding in precision farming and resource management etc. India, as a co-chair of the recently-held Global AI summit, also needs to reassert its push to ‘establish governance and standards that uphold shared values, address risks and build trust’. Besides, AI development must be guided by ethical frameworks that reflect various societal needs and not just corporate and military interests.
Finally, scientists and engineers must embrace Kosambi’s challenge to question the social structures they serve. Artificial intelligence researchers must critically examine the implications of their work, advocating for applications that uplift rather than oppress.
As we celebrate Kosambi’s legacy, his vision urges us to rethink AI’s role in the society. Science, he argued, is not the creation of isolated geniuses but a response to social necessities. Artificial intelligence as the science of our times must transcend its elite origins to serve humanity’s collective needs. This would mean redirecting AI from creating digital divides and fuelling geopolitical rivalries towards solving pressing global problems.
(Mohit Sukhtankar is Assistant Professor in DCT’s Dhempe College of Arts and Science, Miramar. Rahul Tripathi is Professor, Political Science, D D Kosambi School of Social Sciences and Behavioural Studies, Goa University.)
