I presume there is a chance the harm can be undone. As antibiotics are less used, the few resistant bacteria will compete with the many non resistant bacteria and hopefully go extinct.
Extinction is unlikely, unless having resistance comes at a very high cost to the organism. I would not necessarily assume this is the case, though. These are genes that may have already existed in a subset of these bacteria, just as random genetic variation, and then we came along and selectively bred the ones with a natural resistance to our antibiotics.
Think of it like the black death in humans. When this disease suddenly made the jump to humans, some people had naturally higher resistance and some had lower resistance, all just due to random genetic variation. The disease spared people who were more resistant and killed people who were less resistant, and now humans today carry genes that protect us from this disease, even though we have not had another black plague for hundreds of years. Even if we eradicated the disease entirely, we will probably still carry resistance for thousands of years. It's just a part of our genetic makeup now.
Back to bacteria, even if antibiotic resistance does come at a high cost to the organism and there is selective pressure against it in the absence of antibiotics, it is still very unlikely that we can get back to the way it was when antiobiotics were new. Once a deleterious trait appears in a population, it has a way of sticking around (e.g., see numerous deadly genetic diseases carried by humans).
Basically, the cat is out of the bag. We need to drastically reduce our use of modern antibiotics to buy ourselves time to develop new antibiotics--and we need to keep the new antibiotics very tightly controlled.
But my point is that if you cease to apply a selective pressure, suddenly the organism that were suppressed by this selective pressure thrive. Whereas in the case of the back death the pressure didn't disappear, we just developed a resistance (if that's the case).
