Feds must ban development of superintelligence to prevent democratic erosion, says committee witness

The federal government must recognize artificial superintelligence (ASI) as a national and global threat and ban its development in order to protect against erosions of democracy and human extinction, according to a Senate committee witness. Connor Leahy, U.S. director of ControlAI, an organization working to prevent the risks posed by superintelligent AI, told the Senate transport and communications committee on April 15 that ASI not controlled by “trustworthy, democratically elected institutions poses a massive national and international security risk.” Superintelligence is defined as AI systems that can outperform human intelligence and cognitive ability. Essentially, it is technology that is smarter than even the world’s best and brightest. Leahy told senators a major part of the problem is AI researchers and companies “fundamentally do not understand how the AI systems they are creating actually work,” why AI systems make the decisions they do, what their goals are, or how to prevent them from unwanted behaviour (which, he said, they are already demonstrating). This lack of understanding, he argued, means companies do not know how to put effective measures in place to develop the systems safely. While superintelligence does not exist yet, Leahy noted the creation of the first system is widely expected by experts in both industry and academia to be created in the next two to five years, and that it is “rare” to find an expert in the field who believes the systems are more than 10 years away. As several leading AI companies charge forward with their goal of building superintelligence, Leahy urged the government to ban the development of such systems, saying “this stage of the technology poses unprecedented risks, up to and including the extinction of humanity.” According to Leahy, the development of uncontrollable superintelligence could erode democracy in ways that are “hard to imagine compared to today.” He pointed out the effectiveness of mass persuasion through psychological operations run by foreign nation states, adding that AI systems could bring those operations to a “whole new level.” “Every single person could have their own personalized KGB agent assigned to them can convince them of anything,” he said, referencing the Soviet-era security agency.“ can show them fake media, can promise them, can threaten them, can bribe them, can blackmail them.” Leahy gave the committee three recommendations to control the threat of ASI. First, he said, the government must publicly recognize ASI as a national and global security threat, which poses an extinction risk to humanity. Second, Canada cannot act alone. He said it must begin negotiating an international agreement to prohibit the development of ASI. Third, he said, the government should begin closely monitoring the threat of ASI being developed and create detailed scenario planning in order to respond appropriately. Although Leahy recognized Canada has its work cut out for itself in negotiating an international agreement, there is widespread buy-in for the idea, he said. The Statement on Superintelligence is an international effort calling for the prohibition of ASI development and has notable signatories, including Geoffrey Hinton and Yoshua Bengio (two of the ‘godfathers of AI’), Steve Wozniak (co-founder of Apple), and even the British royal family’s Harry and Meghan, Duke and Duchess of Sussex. At the time of publishing, the statement has received over 136 thousand signatures. Leahy said, in his experience, the biggest barrier to garnering support for the movement among elected officials is lack of education. “Most people have never heard of these problems. It’s not that they disagree that this seems really bad, or that we should do something about this,” he said. He added that his colleagues in the U.K. have built a coalition of over 100 members of Parliament that are calling for regulation. He said Canada must be able to form large enough coalitions in order to take action. The support internationally must be enough to overwhelm the leading tech companies, and not allow them to simply defect to other rogue nations. What’s more, Canada must find a way to do all of this, quickly, he said. Leahy argued the leading tech companies are racing toward superintelligence because they are trying to outrun legislation and develop their products before governments can stop them. “On the international front, Canada and other countries should make it a priority to create a ‘trust but verify’ regime where countries commit to not building such technologies in their borders, and enforce this within their borders. And, if necessary, use appropriate sanctions or other means on non-compliance,” he said. When it comes to rogue actors and non-compliance, Leahy argues Canada should reserve the right to self-defense. “The creation of a superintelligence is a direct threat to the lives of our citizens, both in Canada, the U.S., and everywhere else,” he said. “If necessary, we must be willing to take the necessary defensive measures, the same way we would, for example, for threats.” mcollins@thewirereport.ca

Feds must ban development of superintelligence to prevent democratic erosion, says committee witness

Regulatory |
By Maria Collins
|April 20, 2026

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us

Feds must ban development of superintelligence to prevent democratic erosion, says committee witness

Regulatory | By Maria Collins |April 20, 2026

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Ontario appeals CRTC decision on prison long-distance rates

Federal Court judge Phuong T.V. Ngo appointed to Competition Tribunal

OPINION: Canada is building the future in silos — which is why we need a national coordination office

Related Stories

Is what you’re watching AI? You have the right to know, says House committee

Cohere executive won’t confirm company would remain Canadian-owned amid merger rumours

Q&A: B.C. public safety minister talks lawful access, online harms following Tumbler Ridge tragedy

‘Do we need this?’: Artists, musicians and writers question federal approach to AI at national summit

Reuse & Permissions

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us

Regulatory |
By Maria Collins
|April 20, 2026