Ruchi Gupta's Indian Express op-ed names the mechanism the AI-and-jobs debate has been missing: the routine tasks AI now performs are the very tasks through which juniors become seniors. Automate those, keep the seniors to review the output, and a profession loses its first rung. In India that will show up as "never hired", not laid off. Her answer is a national AI and Work Observatory. I have wanted one since 2025; I offer a sharper version, with four quarterly series published as open data.
In this morning's Indian Express, Ruchi Gupta of the Future of India Foundation puts in one line what I have not seen said plainly anywhere (9 November 2026):
The routine tasks that AI can now perform are the same tasks through which inexperienced workers become experienced. If firms retain senior professionals to review AI-generated work while reducing junior recruitment, they may weaken the pipeline that produces tomorrow's professionals.
I argued the middle ground on AI and jobs two years back (Optimism Over AI Job Loss, 20 May 2026). I hold to it. But this argument is different, and it worries me more than any layoff forecast.
Every profession is a ladder whose first rung is routine. A young engineer learns judgement by writing the boilerplate, writing the tests, tracing a failure through logs — precisely the tasks AI now does well. Keep the senior to review the model's output, quietly delete the junior seat, and the work still ships this year and next. What disappears is the supply of experienced people five years out.
In India this will not arrive as layoffs. It arrives as absence. A firm that never posts a fresher vacancy never announces a cut, and no labour bureau records the job that was not created.
The pipeline is already thinning. Nasscom's president Rajesh Nambiar told Nikkei Asia on 6 November, reported two days later, that he would not sugarcoat what awaits graduates:
- About 6 million direct tech jobs in FY26
- 135,000 net additions — 2.3%
- Against roughly 2.6 million engineering enrolments
Rezifi's count of 102,693 open jobs on 969 employers' own careers sites, taken on 8 November 2026, shows where the bottom rung has gone:
- 13,785 openings, 13%, are internships or entry level
- India: 2,073 of 27,659, or 7%, against 14% in the United States
- Indian engineering: 65 of 2,043 backend and full-stack roles, 3%
Quess, reported on 7 November, finds 9.2 lakh engineers listing an AI skill and only 2.57 lakh in true AI roles. Mint reported this morning that cyber-security startups are hiring freshers and training them from scratch for on-site "forward-deployed" roles at ₹18,000–19,000 a month — a bottom rung that survives wherever a client will pay for a human presence.
So when the op-ed asks for a national AI and Work Observatory, I read it as vindication. I proposed a National Employment Grid in July 2025 and made the case again in LMIS vs MAD that September: build the data layer at the source, in real time. I would now narrow the ask to four quarterly series, open data, machine-readable:
- Entry-level recruitment: the entry-level and internship share of live postings, by sector and city
- First-job conversion: of apprentices and interns who start, how many are retained at twelve months
- Wage bands for 0–3 years' experience, so a falling entry wage shows up before a falling headcount does
- AI-tool adoption by sector, so exposure is measured against what firms deploy, not what headlines say
Not a dashboard for Delhi. A dataset for a twenty-two-year-old choosing which course to pay for. I may be wrong about the ladder; a quarterly series would settle it in two years, and I would rather be corrected by my own numbers than be right in a decade of silence.
With regards,
Hemen Parekh
www.Iam-Immortal.ai / www.IndiaAGI.ai / www.YourContentCreator.in
9 Nov 2026
No comments:
Post a Comment