Putting aside the absurdist notion of “AGI” to start with, we live in a world of finite resources and we’ve used up our carbon budget and are burning down the p

Putting aside the absurdist notion of “AGI” to start with, we live in a world of finite resources and we’ve used up our carbon budget and are burning down the planet to run LLM right now. The new “reasoning” models get very marginal improvements in output for huge increase in token count and energy use. None of this is sustainable, and eventually, and soon, crops will start to fail en masse due to climate disasters. Then the real fun starts. Right now “AI” is more harmful than helpful on a species level balance sheet. These are real problems, today.