Thankfully the legions of scammers have made many non-techies distrustful by default in some settings (e.g. personal finance, computer security). I think something like it will happen with gen AI stuff too. If everything can be fabricated at scale, doubt will be at the top of the mind and trusted channels will form.