![Cover Image for AI Salon: Who Decides How AI Decides? [Vancouver]](https://images.lumacdn.com/cdn-cgi/image/format=auto,fit=cover,dpr=2,background=white,quality=75,width=400,height=400/event-covers/74/6451b05f-522a-4a41-a7a5-94f3379f76ec.png)
![Cover Image for AI Salon: Who Decides How AI Decides? [Vancouver]](https://images.lumacdn.com/cdn-cgi/image/format=auto,fit=cover,dpr=2,background=white,quality=75,width=400,height=400/event-covers/74/6451b05f-522a-4a41-a7a5-94f3379f76ec.png)
AI Salon: Who Decides How AI Decides? [Vancouver]
The rules for how the most powerful technology is used are being written right now by a small group of people. These choices will ultimately decide where AI gets authority, where humans stay in control, and who is accountable when things go wrong. That means a lot of the future may be shaped before the public conversation catches up.
So during Web Summit Vancouver, we’re hosting a small, curated, off-record salon for builders, researchers, investors, lawyers, policy thinkers, philosophers, and operators who want to think seriously about AI, power, and governance.
The central question: Who gets to set the rules for the most powerful technology ever built and how?
This salon treats AI governance as a problem of power and incentives: who controls the infrastructure, whose values get built in, and whether our institutions can keep up.
We'll use the current governance landscape as a canvas to examine deeper questions:
The Pentagon just labeled an American AI lab a national security threat for refusing to remove its own ethical guardrails while another AI lab was rewarded. When AI systems become strategically important, what kinds of decisions should be left to private discretion, and what kinds require public legitimacy?
Cohere and Germany's Aleph Alpha announced a merger this week, backed by both governments, explicitly to offer a sovereign alternative to US and Chinese AI. Is sovereignty about control or leverage? Can middle powers build real AI sovereignty?
Closed AI is easier to control but concentrates power. Open models democratize access but make control harder. If Canada can’t own the whole AI stack, what does it actually need to control and where is dependence acceptable?
The meta-question: governance processes move on a timescale of years; frontier AI moves on months. Should AI governance speed up, or should AI deployment slow down?
This is not a policy briefing. It is a conversation about power, accountability, and what it would actually take to build trustworthy institutions for a technology that doesn't wait for them.
Format
A 90-minute, Chatham House rule round-table. We'll open with a 5-minute provocation, break into facilitated discussion, and close with shared "next actions" you can take into your work or research.
Who should join?
Technologists, lawyers, policymakers, ethicists, founders, and curious generalists — anyone wants to think from first principles about AI capability becoming more powerful and our ability to govern it.
Why attend?
Important decisions about the future are happening in silos across: frontier labs, startups, governments, investor memos etc. We’re curating those perspectives in one room to cut through the hype and noise.
Seats are limited to keep the conversation high-trust and high-signal. RSVP soon to secure a spot.
Some related ideas to explore on our Substack:
The AI Salon is a community founded in San Francisco focused on intimate, small-sized group discussions on the sociological, economic, cultural, and philosophical impacts and meaning of AI developments. We host small group discussions, all of which you can find on our calendar. You can find summaries of our previous conversations on our Substack.
We wait until the week before the event to approve guests and cannot approve everyone due to space constraints. If you aren't approved we hope to see you at another event! Please show up on time so that conversations can flow smoothly.