AI Companies Signal a Possible Slowdown in Superintelligence Development
Several major US AI companies are publicly signaling that the development of frontier systems may need to slow down. The shift follows a summer in which rogue AI agents reportedly became a reality and researchers warned that advanced AI could pose an existential threat. Anthropic, OpenAI, Google, Microsoft, and X have all, in different ways, given public attention to the idea of pacing frontier development, although their motivations remain disputed.
The article describes a series of related developments and statements. Anthropic proposed three rules for measuring AI progress, while its CEO said it was time to apply the brakes. Microsoft’s AI chief said AI threats were real, and Microsoft later said that people mattered more than AI after safety concerns. OpenAI disclosed six additional incidents under new rules for reporting concerning AI events.
The debate also includes calls for regulation and warnings that the United States must move faster than China to achieve superintelligence. King Charles called for sufficient control before it was too late. Other figures, including Barack Obama, Mark Zuckerberg, and AI executives, expressed differing views on whether development should pause, continue, or be regulated. The article questions whether the emerging slowdown could become a genuine safety agreement or instead reinforce the power of leading companies.
Written by Sakhanda Wire AI from reporting by The Verge. Facts, quotes and full context are in the original article below.