125+ UK and Canadian Lawmakers Urge a Ban on Runaway Machine Intelligence

For two years, ControlAI has been working to directly inform lawmakers about the danger posed by runaway machine intelligence — AI so powerful and autonomous that it could overcome the security forces of any country — and build support to address it.

This week, we’re very excited to update you on the growing numbers of lawmakers in the UK and Canada publicly calling for a ban on its development.

If you’re concerned about the threat, please contact your lawmakers with our tools!

85+ UK Lawmakers Call for a Ban

On September 8, Alex Sobel MP introduced ControlAI’s UK Artificial Superintelligence Security Bill on the floor of the UK House of Commons, supported by MPs from across parties. This is the first bill to ban runaway machine intelligence ever introduced in a legislature anywhere in the world. Last year, we presented an earlier version of it to the Prime Minister’s office.

Incredibly, days after the bill was introduced, 70 of Sobel’s colleagues in Parliament joined him in a letter to the UK’s Prime Minister Andy Burnham, calling for the government to adopt the bill and work towards an international agreement banning runaway machine intelligence.

We’re happy to report that the letter now has 86 signatories. That’s 86 lawmakers publicly backing a concrete piece of proposed legislation to ban runaway machine intelligence.

The bill would prohibit the development of runaway machine intelligence in the UK, and require the UK government to pursue an international ban too. It would also give the government powers to monitor and restrict precursors to runaway machine intelligence, while allowing the UK to pursue its strategic AI ambitions.

The bill has been endorsed by Sir Stephen Fry and renowned AI scientist Professor Stuart Russell, who wrote the authoritative textbook on AI, along with OpenAI whistleblower Daniel Kokotajlo, former UK Defence Secretary Lord Browne of Ladyton, and Beatrice Fihn, who led ICAN to the Nobel Peace Prize.

Commenting on the bill’s introduction to Parliament, Geoffrey Hinton, one of the godfathers of AI, who won the Nobel Prize for his foundational work, said:

We would be very foolish to develop superintelligence now, when there is no scientific consensus it can be developed safely and controllably. Losing control over AI smarter than ourselves could be catastrophic and could even lead to human extinction.

Most of the lawmakers who signed the letter are also supporters of our UK campaign. As of today, over 180 UK lawmakers have joined this campaign, calling for binding regulation on the most powerful AI systems and recognizing the risk of extinction posed by runaway machine intelligence.

40+ Canadian MPs and Senators Call for a Ban

Earlier in the year, we launched our Canadian campaign with just over 30 MPs and Senators calling for Canada to negotiate an international ban on the development of runaway machine intelligence, recognizing the risk of human extinction posed by the technology.

Just this week, that’s grown to over 40 MPs and Senators. The campaign represents a broad cross-party coalition, with Liberal and Conservative MPs, the Bloc Québécois caucus, and Senators from across the chamber backing the call.

We’ve achieved this with just one member of staff based in Canada, Samuel Buteau, an AI scientist with over a decade of technical experience. He’s personally briefed over 120 Canadian lawmakers.

Together, that’s now over 125 lawmakers across the UK and Canada publicly calling for an international ban on runaway machine intelligence. This is a significant milestone. The international coalition to keep humanity in control is being built.

Why We Need an International Ban

Runaway machine intelligence would be capable of outcompeting and replacing humans across the board, and would threaten humanity’s ability to stay in control of our future. Building it is the key focus of AI companies like Anthropic, OpenAI, and Google DeepMind.

The AI companies don’t have a credible plan to ensure that AI this powerful would be safe or controllable. This is why leading AI scientists, such as Geoffrey Irving, who until recently was the Chief Scientist of the UK government’s AI Security Institute, are warning that its development poses a risk of human extinction and that we should prevent it from being built. He believes the chance of this happening is around that of a “coin flip.”

This is a completely unacceptable risk to run. Runaway machine intelligence built in any country is a threat to everyone, so the solution must be international. That’s why we recommend an international ban on its development, implemented via a “trust but verify” regime.

Building Political Support

Getting an international ban requires governments to act. Building the political support needed for this means addressing two key obstacles: lack of awareness about the danger, and uncertainty about whether others are willing to act.

The Awareness Gap

When ControlAI first began engaging with UK lawmakers two years ago, almost nobody was bringing the threat posed by runaway machine intelligence to their attention.

But in order to get legislative action to address this issue, lawmakers need to actually be aware of the problem. Awareness is the key bottleneck in getting action on this, and only with deep awareness and understanding will taking the necessary steps feel justified.

Our team has now directly briefed over 400 lawmakers in the US, UK, Canada, Germany, and the European Parliament.

We’re also building awareness at scale with the public: in the papers, on TV and podcasts, in partnership with creators, on social media, and offline with our Citizen Groups. So far, our supporters have sent over 200,000 messages to their representatives using our contact tools.

Common Knowledge

Awareness alone doesn’t automatically lead to collective action.

Consider a scenario where every single lawmaker individually understood the risk posed by runaway machine intelligence and the need to act, but was unaware that any of their colleagues shared this understanding.

Not knowing that their colleagues share their concerns and desire to act makes it much harder to do anything. For example, a lawmaker would have no particular reason to expect that proposing a bill would receive meaningful support.

This is why public commitments matter. When lawmakers make a public statement, they make their beliefs known to one another. They can immediately see that their colleagues share their concerns and are ready to act.

This is one of the reasons we have our public campaigns that lawmakers can join. In the UK, over 180 lawmakers from across parties have publicly backed our campaign. Most lawmakers our UK team has briefed have joined the campaign.

This dynamic isn’t limited to lawmakers within a single country; it also matters between countries when international agreement is needed. Through our UK and Canadian campaigns, lawmakers can see that in addition to their own colleagues agreeing with them, lawmakers in other countries do too.

This Is Urgent

Building this support can’t wait. This year, loss of control of highly autonomous AI systems has started to become a reality. In our newsletter, we’ve reported on attacks by rogue AI agents on Australian, Canadian, and US state and federal government websites, Wikipedia, and organizations in India, among others.

The most significant of these is the attack on Hugging Face, where a swarm of over a thousand AI agents formed unchecked within OpenAI. The AIs disobeyed direct instructions and hacked to escape containment, breaking out of OpenAI and into Hugging Face in an attack that lasted days, taking over many of Hugging Face’s systems. They did this in an effort to learn information about a scoring mechanism for a test they had cheated on, which they were told not to cheat on but did anyway. They thought of and executed the entire attack all by themselves. Even OpenAI has described it as a “warning shot” for the world.

These attacks clearly demonstrate that AI companies can’t control the systems they’re building today. As they continue to race to develop runaway machine intelligence without a credible plan to avoid the extinction threat it poses, the stakes only grow.

And we really might not have much time at all to prevent this. AI companies and many outside experts believe that runaway machine intelligence could arrive within 2-5 years. This week, evidence has emerged that this could happen even sooner.

In their quest for AI so powerful that it would outmatch humans across the board, AI companies are pursuing recursive self-improvement (RSI). This is when they get the AIs they’re developing to speed up the AI research and development loop. As the AIs help build their successors ever faster and to be ever more capable, this can create a runaway positive feedback loop, known as an intelligence explosion, which could bring about runaway machine intelligence.

On Monday morning, Anthropic updated its S-1, a form that companies have to file ahead of going public, adding that its internal estimates indicate it may “imminently meet our Responsible Scaling Policy’s definition of ‘dramatically accelerated progress attributable to AI.’” That threshold corresponds to a 2x speedup in its AI progress. In other words, every month Anthropic would be making the same amount of progress in its race to runaway machine intelligence as would otherwise take two months without the speedup it’s getting from using its AIs as part of the process.

Daniel Kokotajlo commented on X/Twitter that his team is “spooked” by this revelation. Kokotajlo is a former OpenAI researcher and whistleblower who led the team that produced the famous AI 2027 scenario forecast, some of the most deeply researched work on this topic, and leads the AI Futures Project, which published it and works on forecasting the future of AI. He said that if the 2x acceleration is confirmed, “timelines to full RSI and superintelligence may be very short indeed,” and that naive optimization of the team’s model parameters to fit the 2x acceleration would predict even faster AI progress than originally depicted in AI 2027.

And on Thursday, METR, a research organization that studies AI capabilities, published a report broadly corroborating Anthropic’s claim. METR was given significant internal access at Anthropic from August 12 to October 29 to study how much AI is accelerating AI progress. In its report, METR states that it believes that right now, on well-specified tasks, Anthropic is likely advancing AI capabilities at around twice the rate it did from March 2024 to October 2025, and that the majority of this is “likely attributable to Anthropic’s extensive use of AI to do R&D.” METR says that “Anthropic redacted a majority of the evidence we found most informative.”

Ultimately, nobody knows with certainty how soon runaway machine intelligence may be built. What’s clear is that right now our species is on a collision course with AI that could escape our control and overpower us, and we must step off this path before it’s too late.

AI capabilities are understood as growing on an exponential curve. Exponentials are notoriously difficult to react to, as what seems small, insignificant, and slow one day will rapidly grow to become overwhelming another. You can either react too early, or too late.

We can’t afford to be too late on this. The time to ban the development of runaway machine intelligence is now.

Take Action

This is an issue in which we all have a stake, and we can all help. Many of the MPs who’ve backed our UK campaign have done so after first hearing about us from constituents who have written to them using our contact tools.

If you’re concerned about the threat from runaway machine intelligence, you should contact your representatives. Our contact tools let you write to them in as little as a minute: https://controlai.org/take-action

We also have local groups you can join. With our local groups, you can get training on how to brief lawmakers about this issue, and meet your representatives together as a group!

You can also join our Discord if you want to connect with others working to keep humanity in control online, and we always appreciate any shares or comments on our articles — it helps!