Below is a video of my opening remarks at Attention: Democracy, Rebuilt. In short, I wanted to frame the event as aspiring to democratic alignment, and to set out what middle powers can do to achieve it. Technical alignment asks whether an AI system does what its developers intend. Democratic alignment asks whether it does what citizens, through their institutions, have decided it should do. By the second standard we are badly misaligned, and the gap is widening. If technical alignment is largely the responsibility of the AI superpowers, democratic alignment is a middle power responsibility. The talk is about what that responsibility looks like in practice. We need to see together, so that what these systems are doing is public knowledge: testing them before they are released, watching how they behave once they are in the world, and understanding what they are doing to our work, our economies and our politics. None of us needs to build that capacity alone. We need to act together, so that public authority sets the terms: the safety of the systems and of the people who use them, the floor any system must meet to enter our countries, and our combined market weight, used to reward the AI we want built. And we need to decide together, so that these choices rest on democratic participation. Governments already work with one another; here they have to work with their own publics, which means citizens who have the knowledge they need, a way to be heard, and governments that answer for the result.

The government is tabling a Digital Safety Act this week. It is expected to be an evolution of last year's Online Harms Act, and this time we are in a very different place. Public support for governance has never been higher, and a new wave of AI products has made the case for guardrails even more urgent. There are four things I will be watching for: 1. Amend the architecture rather than rewrite it. The core model of the Online Harms Act takes the right lessons from UK, the EU, and Australia. It is hybrid approach that creates a strong independant regulator capable of enforcing a duty to act responsibly via risks assessments, transparency requirements and age appropriate design, on platforms (and chatbots - see below). 2. Treat any under-16 access as a pause, not a ban. If the Act includes age restrictions, it must be seen as a temporary measure used to incentivize compliance with the regulations, not as an end in itself. A moratorium that lifts once a platform proves its product is safe puts the burden on companies to fix their design, where a ban leaves it untouched. 3. Bring consumer-facing AI chatbots inside the framework, with obligations built for what they are. Chatbots generate harm directly in private exchanges rather than distributing content that already exists, so the rules should be bespoke to chatbots, not borrowed from social media. 4. Build an innovative regulator with the independence and technical innovation capacity to actually enforce. That means the authority to hire the technical talent the public service rarely attracts and to move closer to the speed of the technology, which a standard institutional template will not deliver. None of this stands on its own. AI is arriving faster, and at greater scale, than the product cycles our institutions were built to govern, and the Digital Safety Act is the first real test of whether Canada's broader AI strategy can keep up. At a moment when trust in government is low, getting this right would be proof that the state can still build important, innovative things, that have the capacity to tackle the harder problems we are on the cusp of. My full argument is here: mediatechdemocracy.com/all-w…

Today I am releasing my submission to the federal AI Strategy Task Force. The government wants Canadians to embrace AI. This is understandable: the technology has genuine potential to improve productivity, reshape public services, and strengthen our economic competitiveness. The challenge is that only 34% of Canadians trust it. The details are jarring: Close to 80% believe AI threatens their job; 70% are concerned about their children's safety; 78% are worried about the integrity of elections. And 88% want stronger governance. This is not a literacy gap. It is a governance gap. This gap between ambition and trust was the focus of my contribution to the Task Force. My report (below) addresses what I think are the core democratic imperatives for any government pursuing AI adoption: ensuring that AI systems are safe; that they support the health of our information ecosystem; and that they are governed in ways that sustain democratic legitimacy. The evidence of democratic harm is accumulating across all three. On citizen safety: AI systems have failed users in mental health crises, and children have been exposed to inappropriate content through AI-powered products. On information integrity: synthetic media is proliferating without provenance or accountability, threatening the reliability of our public discourse and the integrity of elections. On democratic legitimacy: these systems were deployed without meaningful oversight, consent, or consultation, and citizens have no recourse when they cause harm. My submission argues that Canada does not need sweeping new AI legislation to address these problems. We can move now by updating two existing frameworks: → An amended Online Harms Act that brings AI platforms within scope, requiring risk assessments, child safety protections, and data transparency. → An amended Consumer Privacy Protection Act with AI labeling and provenance requirements, so Canadians can distinguish real from synthetic. Both bills died with the last Parliament. We failed this test once with social media, allowing platforms to scale first and govern later. We have a second chance with AI. National AI Strategy Task Force report: static1.squarespace.com/stat… And here is a short video summarizing the core argument made by @MediaTechDem and Paradigms