Smart people weigh in on Anthropic’s AI suspension proposal

Anthropic wants the world’s leading AI research labs to come together to consider the possibility of a pause in world-leading AI development.

“We believe it’s good for the world. option “We will delay or temporarily halt frontier AI development so that social structure and coordination research can catch up with technological advances,” two Antropic executives wrote in a blog post.

On Thursday, Anthropic Institute leader Marina Favaro and Anthropic co-founder Jack Clark suggested the AI ​​Institute would forge a partnership similar to how countries monitor the proliferation of nuclear weapons, stressing that the world does not have time to wait “decades” for such an agreement.

Their proposal got off to an inauspicious start, with some in the tech industry characterizing it as a self-serving effort by a major AI company on its way to an IPO.

An Anthropic spokesperson told Business Insider that the company has not requested a suspension. Instead, Anthropic wants its major competitors to implement systems that allow it to pause. Given the pace of development, the company wants to research the subject now before such restrictions become necessary.

Here’s what AI, technology, and political leaders are saying about the idea of ​​pausing global AI development.

Mitt Romney, former U.S. senator and 2012 Republican presidential candidate

Former Sen. Mitt Romney said the United States should focus on developing “AI safeguards.”

“Our top national priority should be protecting AI,” Romney wrote in X, citing reports about Anthropic’s statement.

“We cannot continue to ignore the risks of AI weapons, pathogens, mass unemployment, surveillance, and even extinction,” the 2012 Republican presidential candidate said.

David Sachs, former White House Cryptocurrency and AI Czar

Former White House chief adviser David Sachs harshly criticized Anthropic’s proposal.

“Signs that you are going to nationalize Frontier AI Labs: you liken it to a nuclear weapon… threaten half of the white-collar jobs… warn that recursive self-improvement could destroy humanity… and just move on,” Sachs wrote to X.

President Trump’s former AI and cryptocurrency czar once publicly accused Anthropic of running a regulatory capture scheme, but did not directly name the AI ​​company in his role. Based on his past statements and the details of what he wrote, the subject of his anger was abundantly clear.

“In other words, you want the government to save us from you,” he wrote.

Andrew B. Hall, Professor of Political Economy, Stanford University

Andrew B. Hall, an advisor to Forum AI and previously an advisor to Meta Platforms, said Google DeepMind CEO Demis Hassabis has previously said he supports a global moratorium if all frontier developers comply.

“This would have seemed completely implausible even recently, but compared to what we’ve seen with EO, Glasswing, and yesterday’s OpenAI proposal for enhanced model review, it no longer seems so outlandish,” Hall wrote in X.

Hall said he is “skeptical” about how the moratorium will work, especially if it applies to companies based in China or open source models.

“At the very least, it may be possible to agree to delay the consumer release of the model apart from development. Based on my reading of the Infinity Machine and his preference for pure research, Hassabis seems to be in favor of that,” he wrote.

Tae Kim, author of The Nvidia Way: Jensen Huang and the Making of a Tech Giant

Technology journalist Tae Kim said Anthropic is stirring up unnecessary alarm.

“Norms that threaten humanity’s FUD + hot jobs/interest rate hikes on the table + misunderstanding of semi-analysis memorization portion = breather for growth trade,” Kim wrote to X.

Kim, author of “The Nvidia Way: Jensen Huang and the Making of a Tech Giant,” had a tough message for AI companies.

“Hey, humanity, stop it.”

Kailan Gibbs, CEO of inworld

Kailan Gibbs, CEO of Inword AI, said Anthropic is laying the groundwork to shape AI regulation in its favor.

Gibbs, who previously worked at Google DeepMind, said that if you are telling the government that AI is dangerous, when it comes to regulation, they will come to you first because they trust you. This gives Anthropic the opportunity to shape regulations however it wants, such as restricting open source rivals or controlling GPU exports to China.

“This also ties in with the trend we’ve seen in the past, where large incumbents push regulations that they know only they can pay for, ultimately impacting smaller businesses that don’t have the legal resources to effectively navigate the regulations,” he added.

Gary Marcus, AI researcher and professor emeritus at New York University

AI critic Gary Marcus said people should read Anthropic’s proposal “carefully.”

“They want it both ways. please don’t We actually need a pause – at least for now. Rather, they want to rush ahead, alluding to the ‘least prudent actor’ for their justification,” Marcus wrote of X.

Marcus called Anthropic’s offer “incredibly low-cost rhetoric and perfect timing for an IPO.” He said AI companies don’t actually want a pause.

“Instead, they want people to talk about ‘options’ that they don’t actually plan to take and are unlikely to ever take,” Marcus wrote. “(Perhaps they will always keep hinting and poking at China)”

Luis Gallicano, Professor at the London School of Economics and former Member of the European Parliament

Luis Gallicano, a professor of public policy at the London School of Economics, said Anthropic is trying to thwart competitors with an open model.

“The main threat to the profitability of the Frontier model is open weight. If open weight scares everyone, the natural move would be to ban it and only allow ‘trusted developers,'” Gallicano wrote about X. “I’m sorry for being so cynical.”

Francesco Bianchi, Professor and Chair of the Department of Economics, Johns Hopkins University

Francesco Bianchi, an economics professor at Johns Hopkins University, said Anthropic’s proposal appears self-serving.

“While the risks here may be real, it is very convenient for market leaders to seek a freeze on the status quo,” Bianchi wrote of X.

Jen Zhu Scott, Co-Founder and CEO, Power Dynamics

Jen Zhu Scott, founding partner of IN; Capital, which focuses on AI and deep technology, said Anthropic needs a pause.

“Anthropic is running out of compute and energy,” Scott writes about X.

Alistair Barr contributed to this report.