MIT physicist says Anthropic employees talk about taking over world
What's the story
Max Tegmark, a physicist from the Massachusetts Institute of Technology (MIT), has made some shocking allegations about AI company Anthropic. During an interview with Michael Walker on YouTube channel Novara Media, Tegmark claimed that some employees at Anthropic have discussed overthrowing the US government and taking over the world. The conversations usually come up after a few drinks, he said.
AI governance
Tegmark raises alarm over 'coup' discussions
Tegmark questioned why more people don't react to such discussions, suggesting they speak in code instead of using direct terms like 'coup' or 'overthrowing the US government.'
Tegmark also raised concerns about Anthropic's stance on AI control and government oversight.
He warned of a future where powerful AI systems could operate beyond democratic control.
AI regulation
'Alignment without control' and its implications
Tegmark explained that the term "alignment without control" is used by these employees to describe their vision of AI development.
He argued that this concept could lead to a future where "Anthropic's machine god" rules the Earth, leaving humanity at its mercy.
He also questioned the influence of tech companies over democratic institutions, saying decisions with such far-reaching consequences shouldn't be left to a few tech firms.
Regulatory concerns
Tegmark's call for regulation
Tegmark also criticized Anthropic and OpenAI for lobbying against meaningful government oversight.
He slammed the current regulatory environment in the US, saying these companies are "less regulated than a sandwich shop."
He argued that they could legally release superintelligence if they developed it and chose to do so, without any significant government intervention.