Why am I not surprised.
As the great statistician George E. Box observed, "All models are wrong, but some are useful."
It has become clear over time that the models climate alarmists rely on are the sort that are wrong and useless.
With advances in AI, I wonder what one could come up with as a model of the earth's climate by requesting that it correctly retrodict regional temperatures, mean global temperature, and sea level, for the past say 10,000 years, taking as inputs, human energy use (not the same as CO2 output, since some forms of fossil fuels burning also produces soot, while, natural gas does not), cosmic ray flux, solar irradiance, land use patterns (forest, agricultural use, city, suburbs,...). It would still be wrong, but it might be more useful than what we've been being fed since the alarmists decided warming rather than in impending Ice Age was the excuse for state control of the energy sector.