I hope so. But I have a feeling the local LLMs will just pick up the slack, running on hardware from liquidated “ai data centres”