It is when Google hasn't found how they want to monetize some of it and more importantly doesn't want their AI to risk brand damage. ChatGPT routinely gets things wrong or semi wrong which is expected with any AI that you have to train with data online and all it takes is their bot to say the right thing that is off base to damage Google's brand image. Google's spent $120 billion on AI in the last 6 years, just look up deepmind and alphaGo and you'll see just how much they've advanced it