OpenAI calls for global standards on safe AI development

OpenAI says the United States should take the initiative in developing global technical standards for frontier artificial intelligence. This comes amid growing concerns among business leaders and researchers that if AI continues to improve on its own in the future, humans may possibly lose control over its development.

The US developer of ChatGPT released a set of proposals on Monday for building international standards for safety and security practices.

It calls for a common technical foundation for capability measurement and evaluation, risk assessment, and safeguard sufficiency. It also underscores the need for a framework for sharing and reporting incidents when AI makes moves that are not intended by humans.

OpenAI says the US should lead in this effort, working collaboratively with countries around the world.

The company warns that without the appropriate care and caution in regard to autonomous recursive self-improvement humans could lose practical control over AI development. It says if that happens, "AI could become more dangerous, less aligned, and, on the whole, a danger to people."

US media outlets are reporting that OpenAI CEO Sam Altman is expected to brief the UN Security Council on Wednesday. Observers are focusing on what will be discussed concerning safe AI development.