Humanity must not rely on AI or face a highly likely loss of its collective expertise

Quantifying the impact of collective AI reliance on collective human brains is a formidable fundamental problem due to each brain's many mysterious neurobiological processes at the molecular, cellular, network, systems, and cognitive levels. By attacking this fundamental challenge, we unify probability theory, tensor algebra, psychology, cognitive science, and neuroscience, and introduce a new theory on the probabilistic loss of humanity's collective expertise due to a collective memory loss. Specifically, we introduce a new mathematical framework on a collective memory loss that we deploy to quantify humanity's probability of losing its collective expertise, whose lower bounds we derive to show that collective dependence on AI maximizes humanity's likelihood of losing its collective expertise. According to this theory, we recommend 15 neuroscience-grounded AI policies that jointly minimize humanity's likelihood of losing its collective expertise.