Paradox of Progress ‘AI and agriculture’

Sri Lanka’s experience illustrates why historical awareness  and inclusive technology deployment are critical for shaping equitable  futures

• If the first revolution made us civilised, perhaps the  second can make us consciously aware of our capacity to choose which  traps we fall into and which futures we create
• Today’s technology offers a second opportunity to shape civilisation, but only if we understand how the first one shaped us The benefits of agriculture were unevenly distributed and often served  emerging elites  more than the majority   
• Without deliberate governance, AI could entrench old hierarchies or create new ones

While the Neolithic Revolution is often celebrated as an  unequivocal triumph, its complex legacy of social hierarchies and  unequal benefits offers a vital lens for our current era. This paradox  suggests that the transformative power of AI is not guaranteed to be  equitable. Understanding the unintended consequences of our first  revolution is key to consciously shaping our second, and ensuring AI  becomes a tool for empowerment rather than a new form of inequality. 
“Today’s technology offers a second opportunity to shape  civilisation, but only if we understand how the first one shaped us.”-  Yuval Noah Harari
Two Revolutions, One Question
Human history is punctuated by revolutions- moments when  our relationship to the world and to each other fundamentally shifts.  Roughly 12,000 years ago, communities in several regions independently  began cultivating plants and domesticating animals.   
This transition, known as the Neolithic Revolution laid the  foundations for villages, cities and states, along with writing,  mathematics and monumental architecture. Now, as artificial intelligence  (AI), quantum computing and the Internet of Things (IoT) transform  economies and societies, many commentators speak of a ‘second genesis’.  Before celebrating or fearing this future, we should reassess what the  first revolution truly meant.  
Agriculture 
The traditional story depicts agriculture as an unequivocal  triumph: a leap from precarious foraging to reliable surplus, enabling  population growth, specialisation, bureaucracies and urban life.  
Yet critics point out its darker consequences.  Jared Diamond called farming humanity’s ‘worst mistake’, noting that  skeletons of early farmers show decreased stature, higher disease rates  and shorter life spans. Yuval Noah Harari goes further, dubbing  agriculture ‘history’s biggest fraud’ because domesticated grains forced  humans into longer work hours and poorer diets.  
Anthropologist Marshall Sahlins described hunter‑gatherers  as living on a ‘Zen road to affluence’ and archaeologist  Mark Nathan Cohen emphasised that health deteriorated when societies  shifted from diverse foraging to monoculture crops.   

Political scientist James C. Scott argues that cereal crops  enabled early states to fix populations in place, tax them and raise  armies, creating domination structures that persist.  
These critiques share a theme: the benefits of agriculture  were unevenly distributed and often served emerging elites more than  the majority.  
 Early Farming
Framing the agricultural revolution solely as a miracle or a  mistake oversimplifies a complex transition. In some regions,  agriculture eventually improved nutrition and life expectancy; surplus  supported cultural and technological achievements from Sumerian  cuneiform to Egyptian pyramids.   
Importantly, hierarchy was not an inevitable outcome of  cultivation. Excavations at Çatalhoyuk, a large Neolithic town in  central Anatolia, reveal no central buildings or temples; residents  accessed dwellings via rooftops and ladders, suggesting there were ‘no  rich and poor nor ruler and subject’.   
Houses were similar in size and layout, and there is little  evidence of centralised authority or social stratification. Such  examples show that early farmers experimented with communal land use and  egalitarian social structures, countering the view that agriculture  invariably produced kings and peasants.  
Regional diversity further complicates the narrative.  Farming appeared independently in the Fertile Crescent, China,  Mesoamerica and the Andes, spreading patchily and interacting with  persistent foraging economies.   
Why did humans delay cultivation for tens of thousands of  years despite possessing modern cognition? Climate, demography and  ecology matter.   
During the last Ice Age, unstable environments made  cultivation risky, while foraging offered dietary diversity, mobility  and social flexibility. Evidence suggests that knowledge of cultivation  long preceded its widespread adoption.   
At the Ohalo II site on the Sea of Galilee, dated to about  23,000 years ago, researchers found wild wheat and barley alongside  dozens of other plant species. In Turkey’s Gobekli Tepe,  hunter‑gatherers erected monumental stone circles over 11,000 years ago,  yet no domesticated plants or animals have been recovered.   
These sites imply that humans could build sophisticated  monuments and harvest wild grains without committing to agriculture;  farming was adopted under specific pressures - population growth,  resource depletion or new social arrangements, not from ignorance.  
Cognitive Continuity, Not Sudden Revolution
Rather than a sudden ‘cognitive revolution’, advances in  symbolic thought and language evolved gradually across hominin species.  Comparative genomics shows that Neanderthals and Denisovans shared the  same functional version of the FOXP2 gene, once thought unique to modern  humans.   
Culture and environment shaped cognition as much as genes.  Modern humans may not have begun farming earlier simply because they did  not need to; many forager societies enjoyed balanced diets and abundant  leisure.  
AI Genesis, Danger or Opportunity
How should we view AI and its siblings in light of this complex history?   
Advocates argue that AI, quantum computing and IoT amplify  human capabilities just as language once did. Precision spraying systems  use AI‑guided cameras to detect and spray weeds, enabling farmers to  reduce herbicide use by up to 90 % and saving fuel and labour.   
Mobile apps powered by AI and smartphone cameras provide  small farmers with on‑the‑spot diagnoses of livestock health issues and  tailor fertiliser recommendations. In education, adaptive learning platforms such as DreamBox  and SmartSparrow analyse student responses in real time to adapt  lessons dynamically.   
Automated grading tools like Gradescope reduce educators’  workload, while AI scheduling software optimises timetables and resource  allocation.  
If ethically deployed, such tools could decentralise  information, automate labour and expand access to knowledge, addressing  some of the inequalities rooted in agriculture.  
Yet technology is never neutral. Without deliberate governance, AI could entrench old hierarchies or create new ones.  
However, the ACLU, a US-based non-profit rights organisation,  warns that AI tools have already exhibited housing discrimination:  Tenant‑screening algorithms trained on biased data have denied people  housing despite their ability to pay, and lenders have overcharged people  of colour by millions.   
Employers increasingly use AI-driven hiring tools that  discriminate against candidates with disabilities or those from  marginalised communities. Because AI is designed and deployed within  systems marked by racial and economic disparities, it can amplify  existing inequities rather than reduce them.  
Sri Lankan Lens
Sri Lanka’s experience illustrates why historical awareness  and inclusive technology deployment are critical for shaping equitable  futures.  
While agriculture no longer forms the economic backbone in  terms of gross domestic product, accounting for less than 10% of  national output, it remains central to rural livelihoods, food security  and socio-political stability.   
A significant portion of the population is still directly  or indirectly reliant on farming, particularly smallholder and  subsistence agriculture.  
The country has made notable progress in digital  penetration; mobile phone usage is widespread, and e-government  platforms have expanded.   
However, disparities in internet access, affordability, and  digital literacy persist, especially in rural and estate communities.  These digital divides could deepen existing inequalities if left  unaddressed.  
AI-driven tools such as precision spraying systems,  smartphone-based diagnostics, and real-time weather and soil analytics  hold immense promise for enhancing productivity and climate resilience.   
If made affordable and supported by training programs, such  technologies could empower Sri Lankan farmers to conserve inputs,  reduce crop loss, and access data-driven decision-making once limited to  industrial-scale agriculture.  
However, deploying AI without deliberate attention to  inclusion risks replicating the very hierarchies and marginalisations  that the Neolithic transition entrenched. For instance, farmers without  formal land titles or credit histories may be penalised by data-driven  lending algorithms.   
Households with poor connectivity may find themselves  excluded from digital marketplaces. Similarly, existing inequalities in  land ownership and gender roles could be magnified if AI adoption is  concentrated among larger or better-connected landholders.  
Sri Lanka’s path forward, therefore, hinges on more than  technological adoption; it demands strategic investment in rural  infrastructure, public-private collaboration, education reform and  equitable data governance.   
Ensuring that AI serves as a tool of empowerment rather  than enclosure will require conscious policy choices. The lessons of the  first revolution warn us that unchecked innovation can concentrate  power; the promise of the second lies in using that insight to design  for justice.  
Recognising both promise and peril, economists, historians  and technologists offer different lenses through which to view our  ‘second genesis’.   
Joseph Stiglitz stresses that innovation does not  automatically deliver shared prosperity; equitable outcomes depend on  institutions, regulations and redistributive policies. Niall Ferguson  highlights that technologies often reinforce existing power structures  before they level them and that historical ‘revolutions’, are rarely  linear.   
Steve Jobs championed human‑centred design and warned  against letting technology dictate values. Elon Musk warns of  existential risks from uncontrolled AI even as he invests in space  colonisation and autonomous vehicles. Jensen Huang of NVIDIA emphasises  the democratizing potential of graphical processing units (GPU) and  open‑source AI, arguing that accessible hardware and software can spread  benefits widely. Their diverse views underscore that tools are only as  transformative as the ecosystems and values into which they are  released.  
Evolution
The real challenge of our second genesis is not to  celebrate AI as a panacea but to decide what kind of civilisation we  want to build with it.   
History shows that surpluses and technologies can both  liberate and constrain. The Neolithic revolution enabled civilisation  and exploitation simultaneously; the AI revolution could do the same.   
Designing social, economic and ethical frameworks that  harness technology for agency, dignity and ecological balance, not  productivity alone, will determine whether AI becomes a new form of  domestication or a path to greater freedom. The great irony is that only  by embracing the very technologies that seem to threaten humanity do we  stand to fully realise it.   
If the first revolution made us civilised, perhaps the  second can make us consciously aware of our capacity to choose which  traps we fall into and which futures we create.
The writer -JSK Senevirathne-is a Senior Lecturer at  the Department of Business Administration, Faculty of Management Studies  & Commerce, University of Sri Jayewardenepura.