#TeamHuman: Keep Humans in Control of AI

The Problem

AI is developing too rapidly for humanity to stay in control.

AI could be the best thing that ever happens to humanity. It could also be the last. A handful of companies are locked in an arms race, building the most powerful technology in human history as fast as they possibly can, with no oversight, no democratic accountability, and, by their own admission, without fully understanding what they’re building.

How many jobs have to vanish?

How much of the internet has to be fake?

How many creators have to be buried under AI slop?

How many artists have to compete with copies of themselves?

How many scams and deepfakes? How many elections undermined?

How many kids have to grow up shaped by algorithms no parent chose?

How close do we get to surveillance dystopias, AI-designed pandemics, killer robots?

Or, as experts keep warning, literal human extinction?

This situation is insane. And we all know it.

The Plan

Slow down. Give humanity time to figure out how to stay in control.

The root cause of all these dangers is the global AI arms race, which creates a race-to-the-bottom on safety and doesn’t give humanity time to figure out how to regulate it effectively.

The Solution

We demand one thing:
slow down AI.

We join the hundreds of Nobel Prize winners, AI experts, and heads of state who signed the Superintelligence Statement, calling for a prohibition on developing smarter-than-human AI until there is 1) broad scientific consensus that it can be done safely and controllably, and 2) strong public buy-in.

The experts agree. Here is how it gets done:

Build the brakes. Frontier AI is built with specialized chips: track and guard them, exactly like we track the ingredients for nukes and bioweapons. Governments set the speed limits on giant training runs.

Enforce the speed limit. An international agreement between the US, China, and major nations: slow down together and enforce real safety standards. The International Atomic Energy Agency verifies nuclear compliance worldwide. We need the same watchdog for AI.

Campaign partner

We’re this close to the finish line.

All four leading AI companies agree things may be going too fast. They say they would slow down if the others did. Recently, more than 1,000 of their own employees signed an open letter calling on the government to help make a slowdown possible. But none of them can stop unilaterally. That’s what governments are for. And governments won’t act without public pressure.

That’s where we come in.
We are the pressure.

We’ve Done This Before

Humanity has come together many times.

Yes, we can slow down, and it would be easy compared to other problems humanity has solved. We can do this.

We limited nuclear weapons

You can’t just buy the ingredients for nukes. They’re tracked and guarded. We can track and guard the chips used to produce frontier AI the same way.

We saved the ozone layer

The world faced a hole in the sky. Countries coordinated, industry adapted, and the ozone layer is healing today.

We eradicated smallpox

A disease that killed hundreds of millions, wiped off the planet by countries vaccinating together. The World Health Organization declared it gone in 1980.

We agreed on rules for war

Even sworn enemies signed the Geneva Conventions. If humanity can agree on limits in wartime, it can agree on limits for AI.

We saved an entire continent

Since 1959, the Antarctic Treaty has kept an entire continent unclaimed, unarmed, and reserved for science.

We restricted thousands of dangerous technologies

Bioweapons, chemical weapons, human cloning: humanity has agreed on bans and restrictions together, thousands of times.

Every time, it took international cooperation, and ordinary people who refused to be silent.
It’s time to do it again.