Part 3/16:
For many years, awareness of AI’s potential dangers grew gradually. Initially, thought leaders like Impolski believed that technological predictions—such as the 2045 singularity—were distant rumors. Over time, as models like GPT-4 demonstrated general capabilities beyond narrow tasks, the perception shifted from a theoretical concern to an urgent societal risk. The explosion in AI safety research—from abstract papers to practical safety measures—reflects this mounting concern.