Yes, my argument is mainly with respect to paying customers who are companies, not individuals.
I have trouble imagining why a company like Anthropic would go through the additional complexity of cheating their individual customers while not doing that to their corporate customers. That feels like a whole lot of extra work compared to just behaving properly.
Especially given that companies consist of individuals, so the last thing you want to do is breach the privacy of a personal account belonging to the person who makes purchasing decisions at a large company!
I mean this earnestly, not snidely - I wish I still had the faith that you do in not being treated abominably by any and every company, or to believe that they wouldn't default to behaving improprerly at any opportunity and for the barest profit margin. It would be nice to still believe that good things could happen under capitalism.
(as a sidenote, I'm very grateful for your insightful and balanced writing on AI in general. It played a considerable part in convincing me to give AI tooling another go after I'd initially written it off as more trouble than it was worth)