Oh that's easy. There's a business model at play here. The point of this information is to make you spend more, and the value of that information is something l

Oh that's easy. There's a business model at play here. The point of this information is to make you spend more, and the value of that information is something like 1%-2% (goods) to 30% (of things like credit card fees you pay) of what you spend more if the shops around you know this information. So from your perspective, the net value of the information is negative. Very negative in most cases. The power of this model is the adversarial aspect. If there's 2 stores, especially online, and store A knows this info it can be 10% more expensive and grow faster than store B. Which means stores can't choose not to pay, it's a losing proposition and these businesses will die if advertising works even a little bit. This aspect is why Google, and everyone, want AI to "fight" over everything: the adversarial aspect. Because for any particular purpose, OCR, or even making entire movies, there's a level of AI that's "good enough". Google or Facebook might reach that level 6 months earlier than everyone else (I would have said 10 years earlier 5 years ago, but even OpenAI's advantage has has shrunk to 2-3 months now, so I feel 6 months is generous), and I'm sure that'll be a good business ... for 6 months, if done well. But if you can create a fight where the best AI wins, then AI is worth the highest amount anyone will offer ...