Again, selective breeding suffers from the same issue of introducing changes that can be detrimental to the organism itself and its place in the balance of the environment. Look at dog breeding as an example. Pugs were bred for a specific look, and that inadvertently caused them to have severe breathing issues. Dachshunds are another example, with many developing spinal issues over time. The difference, as I said before, is the speed; making a change causes unintended side effects - when you make a huge change quickly, it will produce more side effects than making a small change slowly will.

And... again... as I already said... there should be limitations to prevent rolling out new GMOs without specific testing for safety, both in a lab for potential problems to the organism or - in the event of an agricultural product - its consumers, as well as in the environment as a whole, to determine how it may affect the ecology if and when it is introduced. It may take decades to notice changes if the GMO is released immediately after being developed, but if testing protocols are made and followed, we should have no problem quickly spotting any issues before the organism is rolled out into the world.

Just like newly developed medicines need to go through rigorous testing to prevent things like the Thalidomide scandal that caused an immense amount of birth defects due to lax testing, new GMO's will need to be tested as well. But, just like you likely understand the benefits of medicine for helping people suffering from various diseases, GMO's can provide the same level of benefit to people suffering from malnutrition, among a wide range of other positive uses. The key is to study new developments to the point where we can spot and address issues. Throwing away the technology as a whole is not the answer.