Hundreds of Experts Just Demanded a Ban on Superintelligent AI (And 95% of Americans Agree)
A new declaration signed by scientists, former officials, and public figures wants mandatory off-switches on powerful AI and a ban on systems that can improve themselves.
A coalition of hundreds of experts just dropped a document called the Pro-Human AI Declaration, and it's basically a rulebook for how AI should be developed. Spoiler: it's very different from what's happening right now.
The big headline? They want to ban the development of superintelligent AI entirely until scientists can prove it's safe and the public actually agrees to it. Not corporations. Not governments. Regular people.
The declaration was organized by Max Tegmark, an MIT physicist and AI researcher. In a recent interview, he pointed to something pretty shocking: "Polling suddenly is showing that 95% of all Americans oppose an unregulated race to superintelligence."
The document lays out five rules it wants the industry to follow: keep humans in charge, prevent power from concentrating in a few hands, protect the human experience, preserve individual liberty, and make AI companies legally accountable for what their systems do.
Some of the specific demands are bold. They want mandatory off-switches on all powerful AI systems. They want to ban AI architectures that can copy themselves, improve themselves, or resist being shut down. Basically, they're saying: no Skynet.
The timing couldn't be more relevant. The declaration was finalized before the Pentagon-Anthropic standoff, but the two events happening back-to-back has made the conversation impossible to ignore.
As reported by TechCrunch.
Source: TechCrunch
Sponsored