Pragati: AI for Impact- Supporting entrepreneurs to empower people at the last mile

AI enhances last-mile service delivery in justice, healthcare, and agriculture, supported by Adalat AI, Intelehealth, and Farmers for Forests.

India’s development journey has often been shaped by the challenge of reaching those who live at the margins - remote & underserved communities and areas where infrastructure and human capacity are still catching up. The “last mile” is rarely a question of intent or funding alone; it is where gaps emerge despite efforts, because logistical hurdles, social norms, and incomplete data can limit the effect of even well- designed programs. The task ahead is not just to design solutions, but to ensure they are responsive to these on-ground realities, so that progress extends to every corner.

AI's growing significance in India's development is evident as it addresses challenges in underserved areas.

And it is through this lens that AI’s growing role in creating social impact should be understood. Its value lies in going beyond being just a technological fix to supporting human effort - enhancing expertise, reducing burdens, and expanding access where traditional methods may fall short. This was recognised when Niti Aayog’s National Strategy for Artificial Intelligence in 2018 outlined a vision of #AIforAll, focusing not only on the economic impact of AI but also on using it for the betterment of society. The IndiaAI Mission launched in 2025, continues this agenda by emphasising on inclusive datasets, sectoral innovation, and capacity building.

Leveraging AI for the last mile

Let’s visualise this. Walk into a district school in rural India, and the challenges are immediately visible. Teachers manage overcrowded classrooms, attendance is tracked manually, and children often share limited textbooks or resources. The strain is less about intent and more about structural gaps.

This holds true across sectors- judicial courts, health clinics, agricultural farms- where individuals are eager to serve their communities, but systemic barriers often hold them back.

India’s digital public infrastructure, like Aadhaar and UPI, has shown how foundational rails can reshape service delivery. Now, this approach is being replicated across sectors with initiatives like Agri Stack in agriculture which is building a unified digital farmer registry that connects land records, bank accounts, subsidies, and advisories. Similarly, Ayushman Bharat Digital Mission in health, DIKSHA in education, and the Open Network for Digital Commerce for e-commerce are working towards the same.

If designed inclusively, such stacks can ensure that AI solutions connect into trusted population scale public systems rather than work in silos. This possibility is the ethos behind organisations like Adalat AI, Intelehealth and Farmers for Forests, who are part of Pragati: AI for Impact initiative led by Meta and The/Nudge Institute.

Take the justice system. Adalat AI is building a justice tech stack that targets the administrative overload of Indian courts. Procedural delays, paper-heavy workflows, and a shortage of court staff leave judges burdened and litigants waiting. However, Adalat AI’s real-time transcription engine, trained on Indian legal jargon and capable of handling multiple dialects in noisy courtroom conditions, achieves over 95% accuracy in multiple Indian languages. Combined with its case-flow management dashboard that integrates with court APIs to automatically schedule hearings, track case status, and digitise documents, the system aims to cut resolution times for over 4000+ courts. The expected outcome: fewer undertrials waiting in jails, judges with more time for substantive judicial work, and litigants or their families who no longer lose out on livelihoods.

In healthcare, the barriers look different but remain systemic and affect marginalised communities. A large percentage of women delay or forgo seeking healthcare because of barriers like lack of infrastructure, limited mobility, and restrictive social norms. Intelehealth addresses this by equipping nurses with an AI-powered assistant embedded in a telemedicine platform integrated into India’s national digital health stack. The assistant, trained on several rural consultations, offers diagnostic and treatment guidance with 95–98% accuracy. Built to function offline and in 15+ languages, it helps ensure continuity of care even in low-resource settings. Much like how Ayushman Bharat digitally enabled health coverage, this example shows how AI can help extend reach and continuity of care at the last mile.

And in agriculture, the challenge is degraded land and weak incentives for change. Nearly 30% of India’s farmland is degraded, and while farmers know monocropping exhausts soil, transitioning to agroforestry feels risky without trusted information or financial support. Farmers for Forests brings tools to this space: drones and satellite imagery combined with open-source algorithms for tree-crown detection, species identification, and survival tracking. These feed into carbon accounting systems that allow farmers to participate in carbon credit markets with verifiable data. The result: farmers diversifying their income streams, potentially increasing earnings, and restoring ecosystems that sequester f more carbon per acre than conventional practices. Works like Farmers for Forests complements Agri Stack’s vision of linking farmers to markets and government schemes.

Together, these three examples point to a shared lesson. AI at the last mile is about reducing friction where intent already exists, embedding tools into human workflows, and designing systems that that fit real conditions.

Designing for Trust and Inclusion

What distinguishes these interventions is their focus on context. Intelehealth’s AI is refined using rural consultation data, ensuring that recommendations are practical and culturally sensitive. Adalat AI’s tools are tailored to local legal terminology and procedures, making them intuitive for court staff. Farmers for Forests integrates satellite and drone data with traditional knowledge.

Tools like Bhashini are also helping by breaking down linguistic barriers through their AI-driven language platform that provides translation across 22 official languages.

Ethical considerations are equally central. Data privacy, patient safety, and algorithmic fairness are embedded into design frameworks from the outset, guided by legal mandates like the Digital Personal Data Protection Act, 2023 (DPDP Act). Transparent interfaces, explainable recommendations, and feedback loops help foster trust in communities.

Beyond Scale: Precision, Partnership, and Patience

The drive to scale rapidly is common, but India’s digital journey shows that precision and stakeholder feedback are what sustain adoption. UPI and CoWIN succeeded not just because they were designed for scale but also because they were co-created with users, tested iteratively, and built on local trust.

A Broader Vision for Technology in Development

AI could boost India’s GDP by $500–600 billion by 2035. But these gains will most matter if they reach the last mile. AI’s role lies in easing constraints at the grassroots.

For policymakers, funders, and practitioners, the task is clear: design interventions that work with people. By supporting expertise, reducing administrative tasks, and aligning systems with human realities, they can create solutions that are sustainable, scalable and equitable.

The organisations Adalat AI, Intelehealth and Farmers for Forests are part of the Pragati: AI for Impact initiative by Meta and The/Nudge Institute. For more information, visit website.

Note to the Reader: This article is part of Hindustan Times' promotional consumer connect initiative and is independently created by the brand. Hindustan Times assumes no editorial responsibility for the content.