Uh, technically that's true, but the risks are not the same. The first successful kidney removal (nephrectomy) was in 1869. It wasn't for a donation, it was for a woman suffering an uncloseable fistula which, without the operation, would have killed her. So what exactly was she risking? By the time someone was allowed to volunteer to donate a kidney in 1954, nephrectomy had the better part of a century's worth of data and refinement of technique.
Compare this to the risk you're proposing: With current measures which seem to be working, there's a good chance that most people don't get infected. So you're proposing that healthy individuals who would likely not contract the virus, be intentionally infected with a disease which he only have an approximate mortality rate for, and that we have no idea of the long-term effects of.
Surely you can see how the first risk fits withing "First, do no harm", while the risk you're proposing simply doesn't.
That said, I feel like your points are a bit overstated. We now have 66 years worth of Kidney transplant data to show it's safe for both parties, but a lot of people needed to participate in what was a new, risk, experimental surgery to get us to this point.
At some point people have to be willing to take a little bit of risk to get data that could save untold lives and relieve unquantifiable suffering.
Additionally, if COVID-19 does indeed have some horrible long term effect, the cost us having less data or less valuable data now will have a much larger consequence than could ever be suffered by these volunteers.