The use of bots is not to generate new opinions, it is to make fringe opinions seem more popular than they are. Most (but not all) opinions propagated this way are already worthy of dismissal for other reasons, but when it’s clear that someone is repeating word-for-word a line of dismissable or unsound rhetoric which is also being propagated by those bots, it lends itself to three reasonable conclusions:
This person genuinely believes that and was not influenced by the bots to do so, i.e. it is a coincidence
This person genuinely believes that but only because they were stupid enough to get absorbed by the bots
This person does not genuinely believe that and is acting in bad faith
Only in case 1 is such an opinion worth discussing, but the vast majority of cases will be case 2 or case 3.
That is why it is reasonable to dismiss such opinions despite the possibility that they are genuine, in good faith, and not the product of propaganda. Because the odds that they’re not are vastly greater. Nobody can be certain of anyone’s intentions on the Internet, so rational actors can only play a game of “What is the most likely scenario?”.