That is one issue, and the other is end of life disposal of orbiting hardware with fission reactors.

Most of the nuclear powered spacecraft we’ve already launched use a kind of “reactor” called a Radioisotope Thermoelectric Generators (RTGs), which convert the heat from radioactive decay (the fuel is Plutonium-238 or Americium-241) to electricity using thermocouples. These are efficient up to about 100 kW, which is enough for a probe. We have launched over 45 of these things with no problems, but they are tiny.

We have launched only a few fission reactors, with no problems.

The key part of launching a fission reactor safely is that you don’t turn it on. If it’s not turned on, it can’t melt down. The reactor is only turned on once you have orbital insertion. It’s designed to be able to take an impact so if it does fall back to Earth containment should hold…and if it doesn’t, it’s a much smaller reactor that’s not going to cause nearly the problems of Chernobyl. So, it would not pose significant risk unless it released radiation after crashing, in which case you wouldn’t want to go close to it without a rad suit, but again, not nearly the risk of Chernobyl.

For end of life, these craft and stations would be programmed to fly off into space rather than deorbited.

It’s a legitimate concern that is being addressed in the way reactors for space use are being designed.