Decoded: How AI Is Reshaping Work for Women

Abstract

The panel explored how artificial intelligence can be leveraged to expand decent, secure work for women across India’s rapidly digitising economy. Participants examined – the systemic silos that limit women’s access to AI‑related jobs, – the role of “human‑in‑the‑loop” models for creating scalable, gender‑inclusive employment, – government‑led skilling and digital‑university initiatives, – and AI‑driven advisory tools that empower women farmers. Throughout, the discussion stressed the need for demand‑side aggregation, contextual up‑skilling, local‑language capability, and coordinated policy to turn AI’s promise into inclusive growth.

Detailed Summary

A scripted video montage described women’s emerging roles in India’s digital economy (data workers, rural journalists, frontline service providers) while highlighting persistent precarity, low pay and limited career pathways.
Moderator Dr. Sharon Buteau welcomed the audience, introduced the panelists, and asked for a quick visual “photo‑stop.” She then handed the discussion over to Atul Satija.

2. Ecosystem‑Level View – Atul Satija

2.1 The Core Question

“It is not if AI will impact women, but how we embed women into the system so that they are architects, not after‑thoughts.”

2.2 Current Landscape

Job‑certainty: Across global conversations, no clear consensus on whether AI will overall create or destroy jobs; many assume a net loss, while others point to productivity gains for small‑holder farmers and extension workers.
Siloed interventions:
- Government – >22 schemes delivering digital skills to women, but disconnected from actual job opportunities.
- Non‑profits – Skilling occurs close to where people live, not where jobs exist; the approach remains “supply‑side.”
- Industry – Training is often internal L&D (e.g., cappuccino‑making), not a catalyst for new occupations.

2.3 Structural Barriers for Women

Mobility constraints, time‑poverty, household responsibilities limit women’s ability to take up jobs that require physical relocation.
The digital layer of many women’s roles remains invisible, leading to under‑recognition and low wages.

2.4 Recommendation: Demand‑Side Aggregators

Analogy to micro‑finance: Just as last‑mile lenders unlock credit, “last‑mile work aggregators” must connect AI‑driven tasks to women where they live.
Example – Karya (a Nudge portfolio company): Aggregates AI‑first work (data annotation, moderation, etc.) and has built a workforce of ~100 000 women in three years.

2.5 Take‑away Message

Building a travel‑free, demand‑driven platform is the most urgent lever for scaling women’s participation in AI‑enabled work.

3. Private‑Sector Perspective – Mythily Ramesh (NextWealth)

3.1 What is NextWealth?

A social‑entrepreneurial venture that operates “human‑in‑the‑loop” (HITL) services across the AI life‑cycle (data annotation, reinforcement‑learning‑from‑human‑feedback, prompt‑engineering, evaluation).
Presence: 11 centres, ~5 000 staff, 20 000 + jobs created; women constitute ~60 % of the workforce.

3.2 Automation vs. Job Creation

Short‑term: Automation will replace many current BPO tasks.
Long‑term: The volume of AI work will increase dramatically because the proliferation of generative models opens new use‑cases.
Industry forecasts (e.g., Infosys) predict ~92 M jobs automated but ~170 M new AI‑related jobs, suggesting a net gain of roughly 100 M positions and $300–400 B in services.

3.3 Why Human‑in‑the‑Loop Remains Essential

AI is probabilistic → prone to bias and hallucination.
HITL provides validation, quality‑control, and “trustworthiness.” Even when automation covers 50‑60 % of a task, the remaining 40 % still requires skilled human oversight.

3.4 Global Competitiveness & Local Inclusivity

Pillar	Global Edge	Local Inclusion
Speed	Rapid product cycles demand fast up‑skilling.	HITL must be deployed at scale, locally.
Reliability	Trustworthy AI needs human validation.	Female annotators improve model robustness for gender‑diverse contexts (e.g., finance, retail).
Cost	Affordability drives adoption.	Low‑cost local labor keeps services viable.
Business Imperative	Inclusion is a market differentiator.	Women’s participation boosts model accuracy for gender‑sensitive domains.

3.5 Barriers & Operational Solution

Supply‑side: Limited access to firms that need HITL work.
Demand‑side: Lack of awareness that women in small towns can meet volume‑intensive tasks.
NextWealth’s Response: AI‑guided task‑routing, all‑women centres (e.g., medical‑coding hub achieving 99 % accuracy) and a focus on “contextual” annotation that respects local nuances.

3.6 Summary of Insights

HITL work will grow alongside AI.
Investing in women‑led HITL hubs is both a social good and a business necessity.

4. National Skilling & Scaling – Dr. M M Tripathi (NIELIT/MeitY)

4.1 Institutional Mandate

NIELIT is a government body under the Ministry of Electronics & Information Technology, responsible for digital scaling across India.
Physical reach: 56 centres (including 22 in the North‑East, and outposts in remote locations such as Leh, Kargil, Srinagar, Andaman & Nicobar).

4.2 Scale & Impact

Annual throughput: ~1 million learners; ~40 % are women.
Free, industry‑recognised certifications (acceptable for employers in Japan, Taiwan, Korea, Singapore).
Example alumni: Keshriya Sthanao, CTO of Microsoft; Dinkar Gupta, CTO of KPMG Switzerland – both from NIELIT’s non‑formal pathways.

4.3 Digital University Platform (AI‑enabled)

Launched in 4 months; >45 000 registrations; 5 000 registrations per minute at peak.
Program mix (latest): 10 000 AI learners, 3 000 in semiconductor design, plus cyber‑security, AI, and Industry 4.0 tracks.
Features: recommender system, credit‑based assessment, AI‑driven interview simulator, personalized learning paths – all free.

4.4 Bridging Skilling to Employment

Platform directly integrates job postings from 160 industry partners (e.g., Infineon, Intel, Nvidia, Microsoft, IBM).
Virtual labs replicate physical equipment for semiconductor, AI, cyber‑forensics training.
AI‑powered “studio” model upcoming: expert‑led live sessions across 100 studios nationally.

4.5 Innovation Outcomes

28 AI Data Labs delivering community‑driven products:
- Indian Sign Language converter – video/audio to sign language.
- Audio‑to‑audio translation prototype.
Start‑up generation: 25 women‑led start‑ups in the North‑East emerging from NIELIT training.

4.6 Women‑Centric Initiatives

Targeted programs for women returning to work after career breaks (e.g., cyber‑security, AI).
Emphasis on hands‑on, market‑relevant curricula to avoid the “theoretical‑only” pitfall.

4.7 Key Message

Scalable, free, AI‑augmented digital education combined with direct industry pipelines can democratise AI jobs for women in even the most remote regions.

5. AI for Smallholder Farmers – Jona Repishti (Digital Green)

5.1 Problem Context

Smallholder farmers (≈500 M globally) face pest, climate, market‑price volatility.
Women farmers are especially underserved: frontline extension worker ratios of 1 : 1 000, and advice often fails to account for gender‑specific constraints.

5.2 The Solution – Farmer Chat

An AI‑driven advisory chatbot that operates 24/7 in 14 languages across India, Kenya, Nigeria, Ethiopia, Brazil.
Multimodal interaction: users can (a) type, (b) speak, or (c) upload a picture (e.g., leaf symptoms). About 40 % of 8 M queries are multimodal, enabling low‑literacy users to engage.

5.3 Overcoming Access Barriers

Zero‑rating partnership with Safaricom in Africa made the app free of data charges.
Starter‑question library and AI‑generated follow‑up prompts help users overcome “capability overhang” (uncertainty about what to ask).

5.4 Localization & Gender Sensitivity

Fine‑tuning with local annotators ensures advice is context‑appropriate and financially realistic.
Example: a woman in Bihar asked about vaccinating chicks; the AI suggested vaccination, but she lacked funds. The system responded with low‑cost alternatives, illustrating the need for income‑sensitive recommendations.

5.5 Impact Evidence

Reach: >1 M farmers (≈60 % in India) as of the week before the panel.
Adoption: 7 / 10 farmers implement advice within 30 days (survey result).
Women’s experience: Women report feeling safer, more capable, and describe the chatbot as a non‑judgmental space.

5.6 Ongoing Evaluation

Three randomised control trials (Kenya, India, Ethiopia) launched to measure long‑term productivity and income effects.

5.7 Core Insight

AI can bridge information gaps for women farmers, but true livelihood uplift requires simultaneous market linkage and gender‑aware model fine‑tuning.

6. One‑Sentence Recommendations (Rapid Fire)

Speaker	Recommendation (as stated)
Atul Satija	“Policy must recognise women’s distinct constraints (e.g., 40‑hour work week vs. 7‑day week) and adapt labour rules accordingly.”
Mythily Ramesh	“Prioritise building robust local‑language pipelines; no single org can close that gap alone.”
Dr. M M Tripathi	“Create a dedicated fund‑of‑funds for women‑inclusive AI and digital initiatives.”
Jona Repishti	“Envision India as an AI ‘factory’: hubs in cities for development, spokes in small towns for human‑in‑the‑loop work, underpinned by a structured skill ladder for women.”
Moderator (Sharon Buteau)	“Consolidate a national vision that links hubs, skill ladders, and a women‑focused AI ecosystem.”

7. Audience Q&A

7.1 Question (Dheeraj Dolwani – Rural BPO) – Contextual Skilling

Query: How can task‑based skilling (e.g., medical coding) become a game‑changer for women?
Answer (Tripathi):
1. Task‑level scaling – teach core annotation / coding skills.
2. Contextual enrichment – embed sector‑specific knowledge (e.g., local market nuances).
3. Prompt‑engineering & fine‑tuning – train women to refine AI models, moving from rote tasks to higher‑value AI stewardship.

7.2 Question – Adoption of Scaling Courses

Query: How to increase women’s enrollment in scaling programmes?
Answer (Tripathi):
- Overcome mind‑set barriers through targeted confidence‑building programmes.
- Showcase successful women‑led start‑ups (e.g., 25 women‑run ventures from NIELIT labs).
- Emphasise women’s innate strengths in product management and empathy for future‑skill curricula.

7.3 Question – Wage Inequality in Agriculture

Query: Can AI help close the gender wage gap among informal farm workers?
Answer (Jona Repishti):
- Use Farmer Chat to deliver gender‑sensitive advisory that improves yields for women‑managed plots.
- Pair AI advice with market‑linkage services (e.g., price discovery, buyer matching) to translate productivity gains into higher income.

7.4 Closing Remarks

Moderator thanked panelists and the audience, invited Dr. Kapil Sharma to present a commemorative token, and formally closed the session.

Key Takeaways

Demand‑side aggregation is the missing link: Platforms that bring AI‑ready tasks directly to women’s homes (e.g., Karya) are crucial for scaling employment.
Human‑in‑the‑loop work will explode: While automation displaces routine BPO tasks, HITL demand grows with AI volume, creating a net positive job horizon (≈100 M new roles globally).
Local‑language and contextual fine‑tuning are non‑negotiable: Gender‑biased or culturally irrelevant AI outputs undermine adoption; female annotators and local experts improve model robustness.
Free, AI‑augmented digital education can reach the remotest corners: NIELIT’s Digital University platform demonstrates that high‑quality, industry‑aligned training can be delivered at scale without fees.
AI‑driven advisory tools (e.g., Farmer Chat) can empower women farmers by providing 24/7, multimodal, low‑cost guidance, yet must be paired with market access to translate advice into income.
Policy needs a gender lens: Labour regulations, skilling subsidies, and public‑funded AI initiatives must explicitly account for women’s time‑poverty, mobility constraints, and caregiving duties.
Cross‑sector collaboration is essential: Government, NGOs, industry, and academia must co‑design skill ladders, funding mechanisms, and platform ecosystems to avoid siloed efforts.
Evidence‑based evaluation matters: Ongoing RCTs in Kenya, India, and Ethiopia will supply hard data on AI’s impact on women’s productivity and earnings.
Visionary model: Position India as an “AI factory” where urban hubs drive core development while rural spokes host massive, women‑led HITL work, supported by systematic up‑skilling pathways.

Prepared for the AI & Women Future‑of‑Work panel at the Delhi AI Conference, 2026.

See Also:

India AI Impact Summit 2026

Explorer

decoded-how-ai-is-reshaping-work-for-women