Agriculture to health, India reaps AI gains: World Bank
Washington, Aug 10
Artificial intelligence is already saving Indian farmers hundreds of dollars, screening patients for tuberculosis and answering millions of agricultural queries, but it could also threaten call-centre and entry-level technology jobs, the World Bank said in its World Development Report 2026.
The report presents India as a leading testing ground for practical, low-cost AI while warning that countries dependent on business-process outsourcing face pressure as the technology automates call-centre and back-office work.
In Telangana, an AI weather forecasting system helped small farmers modify their behaviour in response to specific risks. Some increased farm spending by as much as one-third, while others recorded net savings of up to $560 per farmer, the report said.
The World Bank said developing economies generally had more to gain and less to fear from AI than wealthier nations. Fewer than one in 10 jobs in developing economies are susceptible to AI automation, compared with more than one-third in high-income countries.
At the same time, one in every six jobs in developing economies-about 16.2 per cent-could be enhanced rather than replaced.
"Developing economies should tune out much of this noise," the report said, referring to fears in richer countries about AI replacing white-collar workers.
India's agricultural sector offers some of the report's most striking examples.
Ama Krushi, launched in Odisha in 2018 and now operating as Krushi Samruddhi, has incorporated AI-powered advisory services. The programme expanded from about three million farmers when it moved into government ownership in 2022 to more than seven million.
A 2025 evaluation found that the service improved farming practices, increased harvests and reduced severe crop losses, producing benefit-cost ratios ranging from 9:1 to 15:1.
Kisan e-Mitra, an agricultural grievance-redress chatbot, provides immediate responses in 11 Indic languages. Within 186 days of adding advanced AI capabilities, it handled about 2.7 million queries-nearly seven times the volume that traditional manual processing could have managed.
The report also highlighted AI's potential to ease India's shortage of medical specialists.
Cough Against TB, developed by Wadhwani AI with India's Central Tuberculosis Division, correctly flagged nine out of 10 people who needed a confirmatory test after screening them through the sound of their cough.
A Government of India assessment in Maharashtra found that computer-aided detection was more accurate and less expensive than readings by radiologists when screening X-ray images for tuberculosis. Tamil Nadu has also introduced large-scale AI screening for diabetic retinopathy, a major cause of preventable blindness.
The World Bank nevertheless cautioned that AI could close off an important path to middle-class employment by threatening call-centre positions and entry-level jobs in software, finance and business services.
It said AI had replaced a capability that customer-service workers at an Indian business-process outsourcing company were already able to perform, illustrating how the same technology could complement workers in one setting but displace them in another.
The report also identified India as one of the developing economies trying to build AI systems adapted to local languages and conditions.
Under the government's IndiaAI Mission, Sarvam AI was selected to develop domestic models. The company plans models with 30 billion and 105 billion parameters, trained entirely in India, according to the report.
India's Unified Payments Interface was cited as an example of digital public infrastructure that allows banks, financial-technology companies and payment applications to use a common underlying system. Such interoperability promotes competition and makes it easier to add or replace providers without rebuilding the entire network, the Bank said.
The report urged developing nations to concentrate on adopting existing AI, adapting it to local conditions and advancing domestic capabilities where resources permit. It said small AI systems could operate offline on phones and laptops or be accessed through voice calls and text messages where electricity and internet connections are unreliable.
World Development Report 2026 is the World Bank's first comprehensive assessment of AI's implications for developing economies. It examines how businesses, governments and individuals are using the technology and where its largest potential gains and risks lie.
The World Bank said AI could enable developing countries to achieve in a decade progress that might otherwise take a century. But it warned that success would require reliable electricity, affordable internet access, local-language data, skilled workers and safeguards against bias, misinformation and excessive dependence on foreign technology.
— IANS
Reader Comments
Impressive numbers from the World Bank report. The Kisan e-Mitra chatbot handling 2.7 million queries in 11 languages is remarkable - that's real scalability. But I'm concerned about the job displacement angle. My company has already started automating customer service roles in Hyderabad. We need better reskilling programs, fast.
The Cough Against TB screening is brilliant! Nine out of ten detection rate through cough sounds alone - this could save millions of lives in rural areas where doctors are scarce. And the Maharashtra radiologist comparison shows AI being more accurate AND cheaper. India leading in frugal innovation, as always!
While I appreciate the progress, let's not ignore the elephant in the room - call centres and BPO jobs. My sister works in a Gurgaon call centre and she's genuinely worried about her future. The report says only 16% of jobs will be enhanced, but what about the other 84%? We need concrete policies for workforce transition, not just optimistic reports.
The UPI example is spot-on - India's digital public infrastructure is genuinely world-class. But the report's warning about dependency on foreign technology is critical. We need to ensure our AI models (like Sarvam's planned 105B parameter one) are truly indigenous, not just hosted on foreign clouds with foreign training data. Atmanirbhar in AI is the way forward!
The benefit-cost ratio of 9:1 to 15:1 for Krushi Samruddhi is phenomenal! But let's also remember - technology alone won't solve everything. We still need better electricity connectivity in
We welcome thoughtful discussions from our readers. Please keep comments respectful and on-topic.