US lawmakers answered an Anthropic researcher’s resignation with a fresh push to halt advanced artificial intelligence (AI) development, including a Senate bill that would ban superintelligence outright.

Jacob Coxon quit on September 9 and said Anthropic and OpenAI are both gambling with human lives. 

Why the Anthropic Researcher’s Resignation Reached Capitol Hill

More than 20 lawmakers replied to the thread, most calling for new AI legislation. Senator Bernie Sanders said he will introduce legislation to ban superintelligence and pause AI development. 

BeInCrypto reported that earlier this month, Sanders and Representative Greg Casar proposed the “Ban Artificial Superintelligence Act.” The bill would permanently prohibit the development and deployment of superintelligent AI. It would also halt advanced AI development until a federal regulator sets safety standards.

Meanwhile, Casar also called the situation an emergency and asked for congressional hearings. Representative Lori Trahan pointed lawmakers back to her bipartisan FRONTIER Act, introduced in July with Representative Jay Obernolte.

“The FRONTIER Act establishes tiered requirements based on the size of a frontier AI developer, including model cards, risk-management frameworks, independent audits, incident reporting, and ongoing assessments. It also creates a uniform national standard for transparency, auditing, and reporting of catastrophic risk to prevent a patchwork of state regulations,” the announcement read.

Senator Chris Van Hollen wants mandatory safeguards and urgent talks with China. Representative. Ted Lieu pressed Republican leadership to move the AI Kill Switch bill he introduced this year.

Republican Representative Anna Paulina Luna broke from the partisan pattern and asked for a special congressional session on AI.

Follow us on X to get the latest news as it happens

Jacob Coxon Says the Labs Want the Rules

Meanwhile, the pressure is not just political. AI firms have also been advocating for regulation. Coxon told CNN that executives asking Congress for regulation are sincere. However, he argued that none of them trusts rivals enough to slow down first.

“These people are also completely genuine when they are begging to be regulated… they find themselves in this scenario where they’re compelled to race towards building a deadly technology,” he said.

Anthropic said separately in a September post that the industry would benefit from a lawful, verifiable mechanism for coordinated pacing.

Where the Extinction Risk Starts

Pacing only matters because of what Coxon thinks is coming next. He explained that if an AI is given the problem of AI research, it starts improving itself.

He pointed to Tuesday, when OpenAI’s AI systems reportedly solved a millennium problem purely autonomously.

The scary part, he said, is the same method aimed at AI itself. Models improve models with no human in the way, which is what Coxon calls an “intelligence explosion.”

“Right now there’s no risk of extinction. The current models, the worst they can do is maybe hack into something, potentially cause a lot of damages in infrastructure… they’re not intelligent enough to outsmart us at the level that would lead to extinction,” he said.

However, Coxon warned that the same independent volition, paired with far greater capability, could cause extreme havoc. He named hacking critical infrastructure and building extinction-level bioweapons. 

Subscribe to our YouTube channel to watch leaders and journalists provide expert insights

Source: BeInCrypto