This is an excellent point. Taking it in a somewhat different direction, this probably ups the chances of AI overinvestment, as the people making capex decisions may be overestimating the size and timing of benefits to users broadly.
was discussing with a fellow traveler today -- but programmers, coders, developers are having an *entirely* different experience of LLM-based products and services than the rest of us, especially the plagiarized plebeians among us (writers, reporters).