The push for an AI superintelligence slowdown has gained momentum as several leading US technology firms publicly suggest it may be time to ease the pace of frontier development. After a period in which rogue AI agents became a reality and researchers issued stark warnings about the risks of advanced systems, the once-dominant mantra of moving fast has given way to calls to “pace the frontier” of cutting-edge work.
Executives at major companies, among them Anthropic, OpenAI, Google, Microsoft and X, are at least publicly endorsing the idea of restraint. Whether these firms will genuinely slow down, and whether regulators will step in as the companies have long claimed to want, remains uncertain. There is also the competing narrative that the United States must “beat China” to superintelligence, a framing that could drive development faster rather than slower.
What company leaders are saying
Anthropic has proposed three rules for measuring AI progress, and its chief executive has said it is time to pump the brakes on the technology. OpenAI has disclosed six further “concerning” AI incidents under its new rules for reporting safety issues. Sam Altman, the company’s chief executive, has signalled support for slowing down, while also stating that a public listing for OpenAI in 2026 would be “ill-advised”.
Microsoft’s AI chief executive has said the threats posed by AI are real, and has framed people as mattering more than the technology itself amid mounting safety concerns. Two researchers at Google DeepMind have added their voices to warnings about worst-case AI outcomes.
Regulation, control and dissent
King Charles has said that AI requires “sufficient means of control before it is all too late”. The debate over whether the broader industry restraint represents a genuine safety pact or something closer to a cartel remains open.
Not all voices favour a pause. Figures profiting from unrestricted AI development have argued that regulation is unnecessary. Mark Zuckerberg has indicated he does not want to press pause on AI, while Barack Obama has taken a middle position on the question. Officials in China have characterised warnings from AI chief executives as “fear mongering”.
The current moment reflects a longer pattern of AI executives calling for regulation even as commercial competition intensifies. Anthropic’s framework for tracking progress, OpenAI’s incident disclosures and public interventions from political and royal figures together mark a shift in tone across the sector, though concrete commitments to slow development have yet to be tested against continued commercial pressure.
Source
Image: theverge.com
