Herman is correct in referring to inductive kickback that will occur in the output transformer of a tube amp in the absence of a load, and Rodman is correct in that damage is unlikely to occur in the absence of an input signal.
However, "unlikely" does not mean that the chances of damage in the absence of an input signal, when there is no load, are zero. Depending on the specific design it is conceivable that transient "signals" which the amp's circuitry may create internally at turn-on or turn-off will induce significant inductive kickback, which could cause damage cumulatively over time.
The most significant concern is not damage to tubes, but the possibility of damage to the transformer itself, due to the extremely high voltages that can be briefly produced when the current through an inductance changes abruptly.
Concerning SS amps, my understanding is also that in most or all cases the absence of a load will not be harmful.
Regards,
-- Al