KI-Unternehmen warnt: Mensch könnte Kontrolle über KI verlieren

Anthropic, a major competitor to OpenAI, has warned that the current pace of AI development is so rapid that humans may soon struggle to maintain control, leading to potential dangers. The central argument is that AI models are becoming increasingly efficient and self-learning, with studies suggesting they can autonomously invent new AI and train themselves significantly faster than humans. While this development is presented as having significant risks for humanity, most companies in the sector lack the incentive to implement development pauses or rigorous ethical regulation, further complicated by the ongoing power competition between the U.S. and China.