Even if India’s number is off by 2 orders of magnitude so the number of cases were actually ~70000 (highly unlikely, since those numbers would show up in hospitalizations), that would 70,000/1.4Bn, so about 1 in 20,000.

NYC at 6000 cases meant it had approx 1 positive in 1000, and the UK at 6000 cases meant it had approx 1 positive in 10,000.

So even if we assume that the US and UK were not undercounting, and that India was undercounting by a factor of 100 (!), India was about half as infected as the UK and an order of magnitude less infected than NYC when it entered lockdown.

India is a pretty big country and just assuming an equal distribution of cases doesn't seem very useful.

The difference between a confirmed case and all cases is that in a confirmed case is a case which has had a viral confirmation. But all cases still involves only those people who have presented for medical attention, because a case is effectively a file about a person. What is unknown is infections.