Today’s WFTD, signal to noise ratio, is the key ingredient in answering this. Lets say you have a musical sound in which there is truly zero energy at 12 kHz. You could put a 50 dB boost on your EQ at 12 kHz and you would still not have any more musical sound at that frequency. You would, however, have added some amount of noise to the signal. How much noise you add would depend upon the quality of the EQ filters and the overall quality of the audio path, but all analog units will add some noise. Your problem in this hypothetical case is that there just isn’t any music at 12 kHz to boost so nothing really happens except you get more noise. I used to see people do this all the time dubbing cassette tapes. The signal to noise ratio of cassette tapes is very low at high frequencies (above 12 kHz). This is due to both the signal being low (poorly recorded) and the noise being greater at those high frequencies. In many cases the noise at 15 kHz is actually louder than the signal. (This is not uncommon in audio. Our ears actually have an amazing ability to discern what is going on even when overall noise is louder than the signal, like on some long distance phone calls.) Boosting these high frequencies enough to hear more hi-end will make the noise that much more obvious, which then reduces the overall signal to noise ratio of the signal. So the reason to be careful boosting frequencies that are low is because you are often boosting as much noise as you are signal.
There are other potential problems with boosting equalizers (distortion, phase shift, reduced headroom, etc.), but most of them also exist when cutting equalizers. There are also situations in which it is appropriate to boost on an EQ. It just depends on the circumstances. The important thing about using EQ is to only use it when you really need to, use no more than you need, and be careful trying to boost frequencies that aren’t there.