Businesses and corporations developing AI are prioritizing profit over preventing human harm

The field of preventing harm from advanced AI systems, spanning alignment, evaluation, and decisions to slow or withhold models whose capabilities outpace their safeguards.