An AI map, if anyone is listening

Although the end of Washington and Anthropic revealed the complete lack of a unified law on artificial intelligence, a coalition of different thinkers has collected what the government has refused to create: a plan of how the development of AI should look.

The Pro-Human Declaration ended last week’s Pentagon-Anthropic standoff, but the collision of these two events was not lost on anyone involved.

“There’s something amazing that’s happened in America in the last four months,” said Max Tegmark, an MIT physicist and AI researcher who helped develop the project. in conversation and this editor. “Sudden polling (shows) that 95% of all Americans oppose unfair competition and intellectual superiority.”

The newly published document, signed by hundreds of experts, former officials, and public figures, begins with the absurd assumption that people are on the way. One method, which the announcement calls “exchange competition,” leads to people being replaced as workers, and then as decision-makers, as power is extended to countless organizations and their machines. Some lead to AI that greatly expands human capabilities.

The latter scenario rests on five key pillars: keeping people safe, avoiding power overload, protecting public affairs, preserving individual rights, and holding AI companies legally accountable. Some of the exercises include restricting intellectual growth until there is scientific consensus that it can be done safely and based on true democracy; approved switches for power systems; and the restriction of architecture that can self-replicate, self-replicate, or refuse to close.

The release of this announcement coincides with a period that makes its urgency easy to recognize. Last Friday in February, Defense Secretary Pete Hegseth designated Anthropic – whose AI already runs on military platforms – a “supportive threat” after the company refused to give the Pentagon unrestricted use of its technology, a symbol reserved for companies with ties to China. Hours later, OpenAI cut its contract with the Department of Defense, which legal experts say will be difficult to use in any way. What was revealed is that DRM’s inaction on AI has been costly.

As Dean Ball, senior fellow at the Foundation for American Innovation, he told The New York Times Afterward, “This is not just a debate about cooperation. This is the first conversation we’ve had as a country about governing AI systems.”

Tegmark came up with a metaphor that most people can understand when we talk. “You don’t have to worry that a pharmaceutical company will release some drug that harms people before they think about how to protect them,” he said, “because the FDA will not allow them to release anything until it is safe.”

Washington turf wars often produce the kind of public outcry that changes the law. Instead, Tegmark sees child protection as a force that can disrupt the current situation. Indeed, the announcement calls for proper testing of AI products — especially chatbots and companion apps aimed at younger users — to cover risks including suicidal ideation, increased mental health issues, and psychosis.

Tegmark said: “If a dangerous old man is texting an 11-year-old boy pretending to be a girl and trying to trick him into killing himself, the boy can be arrested for that. “We already have laws. it’s illegal, so why is it any different if a machine is doing it?”

He believes that once the principle of pre-discharge testing is implemented in children’s medicine, the number will undoubtedly increase. “People are going to come in and be like — let’s add something important. Maybe let’s try if this doesn’t help terrorists build weapons. Maybe let’s try to prove that the intelligence community doesn’t have the power to overthrow the US government.”

It is no small thing that Trump’s former advisor Steve Bannon and Susan Rice, President Obama’s National Security Advisor, signed the same document – along with former Chairman of the Joint Chiefs Mike Mullen and progressive religious leaders.

“Of course, what they agree on is that they are all human,” says Tegmark. “If it comes down to whether we want a human future or a machine future, then they will be on the same side.”