Nvidia says Rubin should have fewer stability problems training with FP4 because of hardware changes - "adaptive compression". There will still be outlier instability inherently, but something they're designing in reduces the cost of managing it.
But yeah, grain of salt - we haven't seen this in practice.
Of course there are techniques such as quantization aware training but I don't understand why a datatype would work for inference but not for that.
You can also abandon backprop entirely but that comes with a whole host of tradeoffs and again why would it work for inference but not for whatever alternative training regime you selected?