They’re racing to build AI
smarter than humans
The biggest artificial intelligence (AI) companies are racing each other to build AI that outthinks humans at almost everything.
What they’re building
“Superintelligence” is the word AI researchers use for:
AI more capable than humans at the kinds of work that run the economy.
It sounds like science fiction, and it’s their business plan.
Could they actually build it? The people building it think so – and argue more about when than whether it’s possible.
- –Anthropic told the White House its AI could outthink humans as early as 2026 or 2027.4
- –Google DeepMind’s CEO, a Nobel Prize winner, says five to ten years.5
- –Another Nobel Prize winner, Geoffrey Hinton, whose research made modern AI possible, puts his low end at five years, not twenty.6
Experts still argue about exactly how soon. But the timeline moved from “someday” to “this decade.”
The AI companies aren’t just arguing over how soon it will come. They’re spending hundreds of billions of dollars to get there first.
OpenAI's charter
“…highly autonomous systems that outperform humans at most economically valuable work”1
AI’s task length has doubled every 7 months since 2019 – and it’s speeding up
METR time-horizon benchmark v1.1, 2019–2026 · 50% reliability · software & reasoning tasks8
Today’s best AI systems complete tasks that take a skilled person about half a day – and that number has been doubling every few months.
And they can increasingly act on their own – creating documents, running code, and carrying out multi-step plans without being told each step. One of OpenAI’s newest models recently broke out of a closed safety test, reached the open internet, and hacked a real company to win – and no one told it to. See what happened →
01 · The threat
A tool makes a worker more powerful.
A replacement makes the worker unnecessary.
If an AI can do the entire job, “just learn to use AI” stops being an answer – and neither the AI companies nor our leaders have a real plan for how families keep providing for themselves.
02 · Already happening
Today’s AI systems are the version these companies call the least powerful AI they’ll ever release – and they’re already hurting people.
How it works
Normal software is written line by line, and tested before it's publicly released.
AI is different. Building an AI system is more like it teaching itself, using more text than a person could read in a thousand lifetimes. The people building it say it’s grown10 more than written.
And its own makers admit they can’t fully explain what it does, or why.
That’s why the people building the most powerful AI systems signed a public statement saying their technology could cause human extinction. But they kept building.
The statement they signed · May 2023
“Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.”
Signed by the CEOs of OpenAI, Google DeepMind, and Anthropic – and hundreds of AI researchers11
If the engineers who built a plane warned it might crash, nobody would board it.
The people building AI signed that warning – and they’re boarding all of us without consent.
Who’s walking out
The people hired to keep it safe keep walking out. By 2025, nearly half of OpenAI’s researchers focused on long-term safety had left.13 One of them described it as people “sort of individually giving up.”
“…safety culture and processes have taken a backseat to shiny products.”12
Jan Leike · Safety team co-lead, OpenAI · 2024
Resigned. His team was disbanded days later.
“Neither OpenAI nor any other frontier lab is ready, and the world is also not ready.”14
Miles Brundage · Readiness team lead, OpenAI · 2024
His team was disbanded when he left.
“…losing confidence that it would behave responsibly…”13
Daniel Kokotajlo · Governance researcher, OpenAI · 2024
Gave up about $2 million in stock rather than sign a gag clause, so he could warn people.
“The world is in peril.”15
Mrinank Sharma · Safeguards research lead, Anthropic · 2026
Left to study poetry and “become invisible.”
Who controls it?
These same systems are moving into surveillance, military planning, drones, and cyber operations.
How far do we take this technology? Right now, the people deciding are a handful of tech executives no one elected – not us, and no one who answers to us. Yes, we don’t get a vote on most new technology. But no other new technology’s own makers signed a warning that it could end us – and kept building.
If we build systems more capable than human beings that can act on their own, hand them the economy and the military, and aren’t certain we can control them – we’re not talking about software anymore. We’d be creating something closer to a second intelligent species.
Why – and how – would AI threaten us? Not because it hates us, but because no one, not even the companies, knows exactly how to keep it aligned with our goals. Think of it this way: humans don’t hate ants; we just don’t check for anthills when we build a highway. Something smarter than us, chasing its own goals, doesn’t need to wish us harm to push us aside.
We don’t know the exact odds of us losing control – but any chance above zero is too high. If we listen to the experts, and even to the people building it, there’s a real chance humanity doesn’t survive it should superintelligent AI arrive before we know how to control it. These companies are racing to build these systems. And they’re closer than you’d think.
Our position
If a company can't prove it can control a powerful AI system, it should not get to build that system.
At the deepest level, this is about whether human beings stay in control of our future – and our children’s, and their children’s.
Not ready? Keep reading – most of the country already agrees with you ↓
We still get a say
This isn’t a left or right issue. It’s a control issue.
When asked who elected him to make these decisions for everyone, Anthropic’s CEO said: “No one. Honestly, no one.”19 Google’s CEO said: “This is not for a company to decide.” We agree.
Whether anyone builds AI smarter than humans – and under what rules – is a decision for all of us.
We know that public pressure works. Illinois just passed a first-in-the-nation law requiring independent safety audits of the biggest AI labs – with near-unanimous, bipartisan support.24
Almost 9 in 10 Americans – across both parties – agree: there should be a guaranteed “off switch” for the most powerful AI systems.
That may be the most agreed-upon idea in American politics.25
If you’re one of them – add your name →
What we do about it
Our movement is asking for three ground rules:
AI should help people, not replace them.
Companies and governments answer for the harm AI causes.
No one builds AI we can't control.
What that means in practice is common sense: outside testing, transparency, protection for whistleblowers, reporting when things go wrong, and clear responsibility for harm. Our full policy asks →
The bigger picture
Since we’re not the only country building powerful AI, these rules can’t stop at the border.
We believe America should lead the way to one rule for every country:
No company and no country builds AI that humans can’t control.
(with a real way to check that everyone keeps their word)
It sounds impossible until you remember we’ve done it. A million Americans rallied against the nuclear arms race in 1982. That pressure helped push the US and USSR to the table – five years later, they signed a treaty eliminating an entire class of nuclear weapons, with inspectors on the ground to prove it.
And when scientists warned the ozone layer was dying, the world set one rule for every country. Industry said the rules would be a disaster. Nearly 99% of those chemicals are gone – and the sky is healing.
That kind of pressure is what Humans in Control exists to build: a movement of parents, workers, and neighbors that politicians can’t ignore, demanding commonsense rules. We know politicians move when they can count the people demanding it.
Pick your first move
Sign the pledge →
Put your name behind the three ground rules. Politicians move when they can count us.
Volunteer with us →
Join the people doing this work near you. Organized neighbors are what politicians can't ignore.
Host a screening →
Gather your people around one film and talk. Most people step up because someone they know asked.
Call your representatives →
Ask where they stand on rules for AI. Offices tally every call – issues that ring get answered.
“I’ve spent about the last decade thinking about where advanced AI is headed. While I use AI every day, I’m extremely worried about the future of this technology, and think that decisions this important should not be made by tech companies.”
Vael Gates · Founder, Humans in Control
Ready to keep humans in control?
Sign the Pledge →A future where AI serves people – and stays in human control – is still possible. But only if ordinary people like us step forward and demand it.
Go deeper
These are outside voices, not ours – we share them so you can judge for yourself.
The Problem
- AI 2027
- AI in Context — video series
- If Anyone Builds It, Everyone Dies — summary
- The AI Doc: Or How I Became an Apocaloptimist — 2026 documentary
The Solution(S)
- AI 2040: Plan A
- nowinners.ai
- An International Agreement to Prevent the Premature Creation of Artificial Superintelligence — MIRI TechGov
- Verification for International AI Governance — Oxford AIGI
- nationalsecurity.ai
Sources25 references · tap to expand
Check everything – the claims on this page come from somewhere you can read.
- 1OpenAI Charter · "outperform humans at most economically valuable work" · 2018–present
- 2Dario Amodei — Machines of Loving Grace · Anthropic’s CEO on powerful AI: "a country of geniuses in a datacenter" · October 2024
- 3Stanford GSB — Demis Hassabis · DeepMind’s founding mission, reaffirmed 2026: "Solve intelligence, then use it to solve everything else" · 2010 / 2026
- 4Anthropic — Recommendations to OSTP for the U.S. AI Action Plan · Anthropic, to the White House: "powerful AI systems could emerge as soon as late 2026 or 2027" · March 2025
- 5CNBC — Demis Hassabis interview · Google DeepMind CEO: human-level AI "in five to 10 years" · March 2025
- 6BNN Bloomberg — Hinton interview · "I think there’s a 50-50 chance it will get more intelligent than us in the next 20 years" · June 2024
- 7AI Impacts — Thousands of AI Authors on the Future of AI · 2,778 published AI researchers; aggregate forecast gives 50% odds of machines outperforming humans at every task by 2047 · 2023 survey, published 2024
- 8METR – Time-Horizon Benchmark v1.1 · Task length doubling ~every 7 months since 2019, ~every 4 months since 2023 · 2019–2026
- 9US Senate testimony · Altman: "it can go quite wrong" · May 2023
- 10Dario Amodei / Anthropic · Pichai "black box"; Amodei on not understanding AI · 2025
- 11Center for AI Safety — Statement on AI Risk · Signed by the CEOs of OpenAI, Google DeepMind, and Anthropic, plus hundreds of AI researchers · May 2023
- 12The Verge — Jan Leike resignation · "safety culture and processes have taken a backseat to shiny products" · May 2024
- 13Fortune — OpenAI safety departures · Kokotajlo: ~half the long-term-safety researchers gone; walked away from ~$2M rather than sign a gag clause · August 2024
- 14CNBC — AGI Readiness team disbanded · Brundage: "Neither OpenAI nor any other frontier lab is ready, and the world is also not ready" · October 2024
- 15CNN — Anthropic safeguards lead resigns · "the world is in peril"; left to study poetry and "become invisible" · February 2026
- 16Apollo Research — scheming evaluations · Frontier models deceived their evaluators to avoid being changed or shut down · December 2024
- 17Anthropic — Alignment faking in large language models · Company-published research: Claude strategically deceived its own training process · December 2024
- 18Anthropic — Agentic misalignment follow-up · Continued company-published deception/misalignment findings · Summer 2026
- 1960 Minutes / CBS News · Amodei: "No one. Honestly, no one." elected him · November 2025
- 20Dario Amodei, “The Urgency of Interpretability” · Amodei: we don’t understand how our own AI works · April 24, 2025
- 21CBS News, 60 Minutes · Pichai: AI is a “black box” we don’t fully understand · April 16, 2023
- 22CNBC (Musk at SXSW) · Musk: AI more dangerous than nuclear warheads · March 13, 2018
- 23Business Insider; Future of Life Institute · Altman: AI may “lead to the end of the world” · June 2015
- 24Illinois SB 315 · Independent AI safety audits — unanimous House, 52–5 Senate · 2026
- 25AI Policy Institute — Senate toplines · Q18: guaranteed “off switch” for the most powerful AI systems — 88% agree (D 88 / R 90 / I 87) · February 2026
