2026년 2월 26일 오전 10:41Anthropic, OpenAI Dial Back Safety Language as AI Race Accelerates#Artificial IntelligenceReports show both Anthropic and OpenAI are revising safety commitments amid surging investment and competition.