AI Observatory
Report or White PaperPublishedNot legally binding
As of October 15, 2026Next listed: 2027 (Proposed pilots in 8 to 10 states)

Roadmap on AI for Inclusive Societal Development

NITI Aayog Frontier Tech Hub roadmap proposing a national mission, Digital ShramSetu, to bring AI to India's 490 million informal workers

India (Union) NITI Aayog (Frontier Tech Hub) October 2025 5 min read
Contents
  1. Key recommendations
  2. Background
  3. What it says about AI
  4. Key provisions or recommendations
  5. Implementation
  6. Current status
  7. Timeline
  8. Sources

NITI Aayog's Frontier Tech Hub released this roadmap in autumn 2025 with an Expert Council and Deloitte South Asia as lead contributors. Instead of focusing on white-collar job loss, it asks how AI and other frontier technologies can serve India's roughly 490 million informal workers. It identifies five core challenges and four systemic barriers, sets out a 2035 vision of voice-first, vernacular AI tools and verifiable credentials, and proposes a national mission, Digital ShramSetu, with a Prime Minister-chaired apex body and a four-phase rollout from 2025 onwards. The roadmap is a non-binding proposal and does not create obligations. The Pacta repository dates it September 2025, while NITI Aayog's publications listing dates it October 2025.

Published. Proposal for a national mission; no sanction of Digital ShramSetu found in sources reviewed

Legal effect: this report or white paper is not legally binding by itself. It may describe or request action; obligations arise only where a statute, rule or court order says so.

Key recommendations

  1. The roadmap centres on India's informal workforce of about 490 million people, who it says contribute nearly half of GDP but work in low-productivity, insecure conditions.1
  2. It names five challenges (financial insecurity, limited market access, lack of skilling, inadequate social protection, low productivity) and four systemic barriers (lack of trust, poor access and usability, low awareness and skills, outdated tools and processes).1
  3. The 2035 vision uses voice-first, multilingual AI assistants, smart contracts and modular micro-credentials designed for low-literacy and offline settings.1
  4. It proposes a national mission, Digital ShramSetu, with a charter, a nodal body, a Prime Minister-chaired apex body, sectoral task forces and state coordination units.1
  5. On AI governance it recommends worker-centric data protection under the DPDP Act, safety, insurance and liability norms for AI and robotic tools, and accountability mechanisms to build trust in AI systems.1
  6. The roadmap sets macro targets for 2035 and 2047, such as raising productivity from US$5 to US$15 per hour by 2035 and social security coverage from 48% to 80%.1

Background

The roadmap was produced by NITI Aayog's Frontier Tech Hub, whose chief architect, Debjani Ghosh, co-signed its forewords with NITI Aayog CEO B.V.R. Subrahmanyam. An Expert Council advised the work and Deloitte South Asia partners were lead contributors.1 NITI Aayog's division reports page lists it as the Roadmap on AI for Inclusive Societal Development, October 2025, under the Frontier Tech Hub.2

The forewords frame the roadmap as a response to AI debates that focus on white-collar job loss. They say little attention has been paid to how AI can serve India's 490 million informal workers, whom the roadmap describes as roughly 90% of the labour force and contributors of nearly half of GDP. It builds its analysis on five worker profiles: a health care worker, a carpenter, a handloom weaver, an electrician and a farmer.1 It describes itself as a living document intended to evolve as new insights and technologies emerge.1

What it says about AI

The roadmap treats AI as one of several frontier technologies, with the Internet of Things, blockchain, robotics and immersive learning. It envisions a 2035 in which voice-first AI interfaces remove barriers of language and literacy, smart contracts guarantee timely payments tied to verified work, and micro-credentials and on-demand learning let workers upskill. It says these tools should amplify workers' skills, not replace them.1

Specific AI uses it describes include context-aware, voice-powered assistants that give guidance through voice, visuals and SMS in regional languages, and a generative-AI knowledge system for artisans, accessed by voice, text, images and video, that draws together loom blueprints, dye recipes, buyer trends and government schemes. It also sketches personalised learning in a worker's own language.1 It recommends scaling vernacular AI through Bhashini and AI4Bharat and funding offline-compatible, lightweight models trained on informal-sector data.1

The roadmap rests on assumptions that include a roughly 80% fall in hardware costs over five years, universal digital access supported by Common Services Centres and kiosks, interoperability by design, and regulatory systems that recognise digital work and regulate AI tools in work environments.1

Key provisions or recommendations

The roadmap offers a set of strategic actions to be carried out through the proposed mission:1

  • Federated trust and credentialing: a federated model through which training providers, platforms, employers and government bodies issue verifiable work and skill credentials, extending IndiaStack with open-source trust and verifiability layers.
  • Innovation incentives: grants, tax breaks, regulatory sandboxes, R&D incentives and public procurement pathways for frontier-technology startups, with pilots in high-mobility sectors.
  • AI and digital infrastructure: vernacular AI through Bhashini and AI4Bharat and offline-capable lightweight models.
  • Knowledge digitisation: digitising informal-sector knowledge into multilingual AI knowledge graphs for industries.
  • Affordable tools and local workforce: local R&D and manufacturing of affordable tools, tech-rental models through self-help groups and trade bodies, and district-level tech operators.
  • Data privacy, AI ethics and user protection: worker-centric data protections under the DPDP Act, edge-compliant platform guidelines, and safety, insurance and liability norms for AI and robotic tools.
  • Grassroots outreach: incentives for states to run mission-mode programmes and to repurpose district infrastructure as digital workforce hubs.

The proposed macro targets include productivity rising from US$5 per hour in 2025 to US$15 by 2035 and US$49 by 2047, social security coverage from 48% to 80% by 2035 and 100% by 2047, and female labour force participation from 15% to 25% by 2035.1

Implementation

The roadmap is a proposal and describes how a mission could be implemented, not a programme already in operation. Digital ShramSetu would have a formal charter, a nodal body that tracks metrics through dashboards, an apex implementation body proposed to be chaired by the Prime Minister with ministries such as Labour and Employment and Skill Development represented, capability teams, sectoral task forces and a state coordination unit.1

It proposes four phases: mission orientation (2025 to 2026), when a mission charter would be drafted; institutional setup and governance design (2026 to 2027); pilots and select programme launches (2027 to 2029), including pilots in 8 to 10 states; and nationwide rollout from 2029. A multi-level KPI framework would run from Phase 2.1 The roadmap lists strategic, financial, technology, operational, intellectual property, supply chain, and data and cybersecurity risks to be addressed in the charter, and does not give a budget figure in the sections reviewed.1

Current status

As of 15 October 2026 the roadmap is available on NITI Aayog's website under the Frontier Tech Hub.2 The sources reviewed for this entry did not show that the Government has sanctioned or launched Digital ShramSetu, set up the proposed apex body, or issued a mission charter. The publication date differs between sources, October 2025 on NITI Aayog's listing against September 2025 in the Pacta repository.

Timeline

  1. Sep 2025

    Date given in repository

    The Pacta repository records the roadmap as September 2025; NITI Aayog's own listing says October 2025.

  2. Oct 2025

    Roadmap listed by NITI Aayog

    NITI Aayog lists the Roadmap on AI for Inclusive Societal Development under its Frontier Tech Hub reports.

  3. Oct 2025

    Proposed Phase 1: mission orientation

    The roadmap proposes drafting a mission charter in 2025 to 2026.

  4. 2026

    Proposed Phase 2: institutional setup

    Governance structures and the implementation blueprint would be finalised in 2026 to 2027.

  5. 2027

    Proposed Phase 3: pilots

    Pilots, including in 8 to 10 states, would run from 2027 to 2029.

  6. 2029

    Proposed Phase 4: nationwide rollout

    Proven solutions would be scaled across the country from 2029.

Sources

  1. [1]
  2. [2]
  3. [3]
    NITI Aayog home page
    PrimaryNITI Aayogniti.gov.in
Report an error Revision history
Cite this entry

Pacta AI Observatory, “Roadmap on AI for Inclusive Societal Development,” last updated October 15, 2026, https://aiobservatory.pacta.in/wiki/niti-aayog-ai-inclusive-societal-development-2025.

Facts verified October 15, 2026 · Last edited Oct 15, 2026 · 12:37 Not legal advice. Check the primary sources before relying on this summary.