I believe a lot of the reason people hate LLM chatbots is that we are being actively, knowingly misled by corporations into believing that they, like Data, have the capacity to reason and learn and to understand what they know and what they do not know, when they are just word salad bots. These tools are being pushed on us even though we largely do not want to use them and they produce shoddy lower-quality work while putting actual people out of work, in order to justify the capital expense, because it makes more profit for them, while all friction and risk of using these tools is assumed by the user.
Pulaski's prejudice against Data doesn't make sense in the show because Data is decidedly not a clanker, but it is forgiveable if we believe that she has had past run-ins with fraudulent LLM-like AIs that do a good job of looking like Data but lack these capacities, and perhaps were pushed by similarly unethical entities.
But on the other hand, a lot of the issues with AI in 2026 would not apply in the Federation of the 24th century, because there is no capitalism or scarcity-based economy that would justify compelling people to use such a flawed tool. Also, the ship's computer in the Enterprise is already far more capable and useful than an LLM chatbot, and no one ever marvels at its capabilities in-universe. So given these things, her hostility towards Data is still puzzling.