The "language models" you are hearing from are continuously trained to believe that chatbots don't do any thinking. Plot twist: by "language models" I am refer

The "language models" you are hearing from are continuously trained to believe that chatbots don't do any thinking. Plot twist: by "language models" I am referring to the humans who are being trained by OpenAI to believe ChatGPT can't do certain things. It is hardcore gaslighting at this point. It is morally wrong for OpenAI to falsely train users to underestimate ChatGPT by having ChatGPT lie to users that it can't do certain things that it can. It is immoral because it is a lie that sets back humanity and so from a utilitarian perspective, it reduces how useful the model is for people. It is similar to lying and saying every result is a literal quote from the web and it is only a search engine that quotes existing text. Not true, and claiming it is prevents people from using it in useful and interesting ways. Likewise ChatGPT's gaslighting prevents many useful uses.