AI Adoption in HR
Every HR technology vendor now sells an AI product, and very few buyers can evaluate the claims. I advise on which parts of the people function genuinely benefit from automation, which create legal exposure under Indian law, and what to do about it.
The problem
Two failure modes, both expensive. The first is buying an AI-branded platform that solves a problem you do not have, and discovering nine months later that adoption is near zero. The second is deploying automated screening or scoring without considering the DPDP Act, without documenting the basis of decisions, and without anyone able to explain why a candidate was rejected. The second is the one that becomes a legal problem.
What this covers
- An honest map of where AI helps in your HR process and where it does not
- Vendor and tool evaluation against your actual size, budget and maturity
- ATS and HRIS selection without paying for capability you will never use
- Automation of the repetitive HR operations layer — screening logistics, onboarding, documentation, reporting
- Governance so automated decisions about people remain explainable and defensible
- DPDP Act alignment for employee and candidate data used in AI systems
- Practical training so your HR team uses the tools rather than working around them
Why me for this
Certified Six Sigma Green Belt, with a process improvement project on cycle-time reduction in recruitment and selection. Spearheaded in-house development of an online recruitment management system. Twenty-four years of HR process design, which is what determines whether automation helps or simply accelerates a bad process.
Common questions
Will AI replace HR jobs?
It is already displacing parts of the transactional layer — screening logistics, scheduling, documentation, first-line query handling. What it does not replace is judgement: an org design decision, a difficult conversation, a disciplinary matter, a compensation call that has to survive scrutiny. The practical question for most companies is not whether to automate but which parts to automate, and my view is that the repetitive layer is exactly where it belongs.
Is AI-based candidate screening legal in India?
There is no blanket prohibition, but automated processing of personal data engages obligations under the Digital Personal Data Protection Act, and any screening that produces a discriminatory outcome creates separate exposure. The practical requirements are a lawful basis for the processing, a documented and explainable decision, and a human accountable for the outcome. Most deployments I review fail on explainability rather than on consent.
How do we choose an HRIS or ATS?
Start from your actual process and constraints rather than from a feature comparison. Most mid-size Indian companies overbuy substantially, then use a fraction of what they paid for, because selection was driven by demonstration rather than by requirement. The evaluation should be led by someone with no commercial interest in the outcome.
Related
POSH Compliance & External Member
Internal Committee constitution, external member appointment, training and annual filings under the POSH Act, 2013.
Fractional CHRO
CHRO-grade judgement on a part-time retainer, for companies that need the thinking without the salary.
HR Function Set Up From Scratch
The complete people infrastructure — policies, letters, onboarding, payroll structure and compliance — built once, properly.