CEOInterviews.AI
Start App
Aidan Gomez
CEO and Cofounder, Cohere

AI regulation 'needs to be led by the public sector': One-on-one with Cohere CEO Aidan Gomez

📅 Sep 15, 2026 CTV News 10 MIN 317 VIEWS 17 SEGMENTS · 2 SPEAKERS
Cohere CEO Aidan Gomez speaks about the dangers AI poses after a recent article published by the CEO of Anthropic calling for the alleviation of antitrust laws. Subscribe to CTV News to watch more videos:    / ctvnews   Connect with CTV News: For live updates and latest headlines visit: http://www.ctvnews.ca/ For breaking news, fast, download the CTV News App: https://www.ctvnews.ca/app Must-watch stories and full programs at http://www.ctvnews.ca/video CTV News on TikTok: https://www.tiktok.com/discover/CTV-News CTV News on X (formerly Twitter):   / ctvnews   CTV News on Reddit:   / ctv...

What Aidan Gomez said

Written from the verified transcript and checked against it. Every figure links to the moment it was said.

Aidan Gomez, CEO of Cohere, argued that AI regulation should be led by the public sector, not by a small group of Silicon Valley companies. He criticized the Anthropic CEO's call to relax antitrust laws, calling it a red flag and a potential case of regulatory capture. Gomez acknowledged real risks like cyber threats and bioweapons but dismissed extinction scenarios as fiction. He advocated for specific, measurable, and testable regulations. On data centers, he emphasized their necessity, debunked misinformation about water usage, and highlighted Canada's clean energy advantage. He urged Canada to become a data center superpower, leveraging its AI heritage from Canadian researchers like Hinton and Bengio, and stressed the importance of community consultation and sustainable practices.

Key takeaways

  1. AI regulation must be led by the public sector, not by a small group of Silicon Valley tech companies.
  2. The Anthropic CEO's call to relax antitrust laws is a red flag and could enable regulatory capture.
  3. Canada should become a data center superpower, leveraging its over 60% clean energy grid.

Numbers and commitments

FigureWhat it refers toTypeAt
60% Clean energy share of Canada's grid metric 8:21

Chapters

  1. 0:00Criticism of AI safety cartel
  2. 1:48Regulatory capture concerns
  3. 3:02Public sector leadership in regulation
  4. 4:02Dismissing sci-fi extinction risks
  5. 5:20Data center development and misinformation
  6. 8:21Canada's clean energy advantage

Questions asked in this interview

6
  1. 0:03Can you expand on that for me?
  2. 1:48Do you think they're trying, if I could be blunt, to leverage concerns around safety for their own commercial interests?
  3. 3:02Do you think there's a regulatory framework that the public sector could come up with that allows both those things to happen at the same time?
  4. 4:02So why do you think that isn't real?
  5. 4:42You're not worried about the risk of that?
  6. 5:20How can those concerns be mitigated or addressed in a way that still allows Canada to be what you envision in that space?
Interviewer 0:00 ↗
Hi, welcome to our program. Thanks very much for making the time for this.
Aidan Gomez 0:02 ↗
Yeah, thank you for having me.
Interviewer 0:03 ↗
I appreciate it. I do have a lot of questions about investment, the summit, data center, stuff like that. But I have to start on the post I read of yours. I'm sure you're getting asked about it non-stop as a follow to sort of what we're hearing from some of the bigger American companies about the prospect of really the dangers that in their view AI presents and they're having this call to slow down their own development. You are not in favor of that and you labeled it kind of as a cartel. Can you expand on that for me?
Aidan Gomez 0:28 ↗
Yeah. Well, the first thing to say is that there are dangers to the technology for sure. In particular, things like the cyber threat like this is the most capable cyber weapon we've ever seen developed. So, that is 100% real and that threat needs to be mitigated. There's a second category of risk where people talk about extinction or the Terminator like scenarios that are kind of sci-fi inspired and they even like to assign percentage likelihoods for that to happen. Those aren't based in fact. Those are fiction and they shouldn't be used, they certainly shouldn't be used to motivate policy. With the particular article that was put out by the Anthropic CEO over the weekend, it called for the alleviation of antitrust law so that antitrust law that was set up to prevent the formation of cartels specifically for his company and a couple others so that they could better coordinate amongst themselves. That should raise a huge red flag. You know, I am one of the CEOs of these AI companies and yes, I hope that we're listened to when it comes to regulation. I hope we do have a voice, but there is a much broader set of stakeholders that need to be consulted and this needs to be led by the public sector, by the government, not by a small group of Silicon Valley tech companies.
Interviewer 1:48 ↗
Do you think they're trying, if I could be blunt, to leverage concerns around safety for their own commercial interests?
Aidan Gomez 1:54 ↗
I don't want to presume their incentives or intentions but certainly you could imagine a scenario where the companies who have reached a certain level of capability in AI want to pull up the ladder behind them and not let others come up and compete. It's called regulatory capture. It's been used many times in the past. And that's something that you know we all as a public have an interest in defending against. This needs to be led by the government. That's the best path towards safe AI deployment. The other thing is that there's so many benefits to come from this technology whether it's reduced cost of living for the average Canadian by the efficiency gains that can be boosted from companies adopting, the economy adopting AI or new drugs and cures for different diseases. We're starting to see the fruits of this technology play out. So while we absolutely need to mitigate the downsides, we need to run towards the good, right? The life-saving, the things that actually materially impact our economies and the individual.
Interviewer 3:02 ↗
Do you think there's a regulatory framework that the public sector could come up with that allows both those things to happen at the same time?
Aidan Gomez 3:09 ↗
100%. Of course. Yeah. There's good regulation like unequivocally. Right. There's regulation that doesn't entrench the incumbents, supports innovation but sets clear guardrails in the places where there is concern and I put out a blog post that talked about the different areas that I think we need to invest. We need to identify what the risks are, what are we mitigating against because the sci-fi scenarios aren't specific enough. So we need to get down to details of what we're trying to protect against, we need to be able to measure those behaviors and then we need to have testing to assess these models for those risks. So those core components are something that we can implement as a government. We can empower regulators to do that. And I think it's a much safer path than giving the power to some of my peers in Silicon Valley.

9 more exchanges in this transcript

Sign in free to read the rest of this interview. No card required.

Sign in to read the full transcript

Cite this transcript

APA, MLA, BibTeX
APA

Gomez, A. (2026, September 15). AI regulation 'needs to be led by the public sector': One-on-one with Cohere CEO Aidan Gomez [Interview transcript]. CTV News. CEOInterviews.AI. https://ceointerviews.ai/interview/1352096/

MLA

Aidan Gomez. "AI regulation 'needs to be led by the public sector': One-on-one with Cohere CEO Aidan Gomez." CTV News, 15 Sep. 2026. Transcript, CEOInterviews.AI, https://ceointerviews.ai/interview/1352096/.

BibTeX
@misc{gomez2026_1352096,
  author       = {Aidan Gomez},
  title        = {AI regulation 'needs to be led by the public sector': One-on-one with Cohere CEO Aidan Gomez},
  howpublished = {Interview transcript, CTV News. CEOInterviews.AI},
  year         = {2026},
  month        = {sep},
  url          = {https://ceointerviews.ai/interview/1352096/},
  note         = {Speaker-attributed transcript with timestamps}
}