I do wonder what models it uses under the hood. ChatGPT already knows more about me than Google did before LLMs, but would I switch to inferior models to prese

I do wonder what models it uses under the hood. ChatGPT already knows more about me than Google did before LLMs, but would I switch to inferior models to preserve privacy? Hard tradeoff.