Building standards for the next phase of AI

OpenAI calls for U.S.-led international standards on frontier AI safety and recursive self-improvement.

OpenAI argues that alignment research must keep pace with more autonomous systems, including automated AI researchers that could accelerate recursive self-improvement. The post says fully autonomous RSI is not happening today and should not be pursued until human control can be preserved, citing a previously disclosed Hugging Face incident as a preview of sharper risks. It warns that conflicting national evaluations and incident definitions would fragment evidence and weaken collective oversight of both open and closed models. OpenAI says the United States should lead international technical standards with bodies such as the Center for AI Standards and Innovation and national AI safety institutes.