Indian industry should help write the rules for AI, early and in the open

Every major economy is now writing rules for artificial intelligence. Some will be strict, some permissive, most somewhere in between. India will write its own, and it will also live with rules written elsewhere, because Indian companies sell to customers in Europe, North America and beyond.

My argument is simple. Indian industry, especially the companies that deploy technology in the field rather than only building it, should take part in that rule-making early and openly.

Why deployers matter

Rules are often written with the model builder in mind. But most of the risk, and most of the benefit, shows up at deployment: in how a system is configured for a bank branch, how it is supervised in a hospital, how it is maintained in a school. The people who do that work know where things go wrong. They know which safeguards are practical and which exist only on paper.

If those people are absent, we get rules that are either too vague to follow or too rigid to work in Indian conditions.

Early, and on the record

Taking part does not mean lobbying behind closed doors. It means publishing what we have learned, responding to consultations in writing, and being willing to have our positions examined. Public input builds trust in the final rules, and it holds industry to what it says.

Rule-maker, not rule-taker

India has a chance to influence how AI is governed globally, as it has with digital payments and identity. That influence will come from evidence: what we have deployed, at what scale, with what results. Indian companies can supply that evidence. They should.