I feel its similar to Microsoft asking for 'goodwill'.

No, sorry, you've used up the goodwill; the goodwill-bank-account balance is empty.

I believe --- that Meta and Google haven't proven themselves trustworthy.

I believe that the only way AI can be truly trustworthy is for it to be open source and run locally --- possibly behind a filtering firewall.

Does anyone really want their private, personal data to be used against them?

These companies were built on privacy invasion and I see no reason to believe their AI won't be an extension of this.