The bait-and-switch critique is valid, but the real pragmatic issue is that AI companies are discovering their unit economics don't support flat-rate pricing fo

The bait-and-switch critique is valid, but the real pragmatic issue is that AI companies are discovering their unit economics don't support flat-rate pricing for compute-intensive services. Try running AWS on a Netflix subscription model. The transparency problem compounds this. The sustainable path forward likely involves either much more transparent/clear usage-based pricing or significantly higher flat rates that actually cover heavy usage.