OpenAI, the developer of ChatGPT, has urged that the United States should take the lead in establishing next-generation artificial intelligence (AI) technology and safety standards.

In a post published on its official website on the 21st local time titled "Building Standards for the Next Phase of AI," OpenAI stated, "We believe the United States must lead efforts to establish technological standards for cutting-edge AI in cooperation with countries around the world."

OpenAI pointed out, "Whether we seize leadership now will determine whether the U.S. leads the global AI architecture or watches a fragmented, imbalanced, and conflict-ridden system take root from the sidelines."

It argued that the U.S. AI industry is currently at the technological forefront and that the nation holds an advantageous position in key areas such as finance, trade, defense, technology, and intelligence, placing it in a strong position to secure leadership.

"Geopolitical competition is not simply a matter of who is ahead technologically, but also a competition surrounding the adoption and diffusion of AI," OpenAI emphasized, adding, "In particular, the realm of AI adoption and diffusion will be determined by who leads cooperation and rational rules."

As a way to establish such international AI standards, OpenAI suggested utilizing a network composed of AI safety institutes already established in various countries, mentioning South Korea as one of the 10 example nations that have built such institutes.

Mentioning the scheduled summit between the U.S. and China, OpenAI commented, "If dialogue between the U.S. and China takes place in this field, it will be a positive development," adding, "The upcoming summit is being held at a timely moment."

Meanwhile, regarding the debate over "slowing down AI," which has recently been at the center of controversies in the AI industry, OpenAI added, "Slowing down the pace of AI development does not mean setting a predetermined speed and maintaining it," but rather "ensuring that 'alignment' research and its deployment stay ahead of AI capabilities."

Seemingly mindful of the concentrated concerns over "recursive self-improvement" (RSI), where AI designs AI itself, the company emphasized, "Fully autonomous RSI is not being realized at this time," and "it should not be pursued until it can be implemented safely."

Chief Executive Officer (CEO) Sam Altman is scheduled to attend a United Nations Security Council meeting directly on September 23 to brief on the topics of AI and international security.
 
(Photo: Getty Images)