Inside Canada’s $240M Sovereign AI Push at Web Summit Vancouver

The lines of control in the artificial intelligence gold rush are being aggressively redrawn, framing an escalating clash between closed proprietary ecosystems and open-source replication that could dictate the next decade of tech governance.

Opening night of Web Summit Vancouver drew more than 20,000 attendees from over 100 countries, serving as a highly politicized battleground for the future of digital sovereignty.

“A battle is raging for the future of AI,” Web Summit CEO Paddy Cosgrave told the crowd at the Vancouver Convention Centre, zeroing in on the structural friction point dividing the industry: whether AI should remain behind the guarded walls of well-capitalized tech giants or be distributed openly to the masses.

The Cost of Replication and the Licensing Friction

As foundational model developers guard their IP with increasingly litigious fervor, the technical reality of the market is shifting. The conference spotlighted the diminishing moat of closed-source systems through a highly discussed case study by tech builder Sigrid Jin.

Jin detailed the on-stage reconstruction of Anthropic’s Claude codebase—a massive feat achieved at a reported cost of 25 billion tokens.

The breakthrough highlights a looming crisis for Big Tech’s business models: if multi-billion-dollar proprietary models can be effectively replicated through raw compute and token consumption, the industry’s economic battleground shifts from basic architecture to model provenance and dataset licensing. Jin warned that this emerging “agentic era” of autonomous AI will fundamentally disrupt current copyright structures and forced a rewrite of standard licensing norms.

Canada’s Bid for Digital Sovereignty

The corporate ideological battle is unfolding alongside a aggressive geopolitical push by the Canadian government to anchor itself as a global AI counterweight.

British Columbia and federal officials deployed a combined marketing and fiscal blitz to capture incoming capital. The provincial government has committed $6.6 million over three years to secure the Web Summit footprint, aiming to double attendance to 40,000 by 2027.

More critically, Canada is attempting to operationalize the concept of “digital sovereignty”—the idea that nation-states must control their own data infrastructure and compute capacity rather than outsourcing it to foreign tech conglomerates.

The country’s newly appointed, first-ever Minister of Artificial Intelligence and Digital Innovation, Evan Solomon, took the stage alongside Cohere Chief AI Officer Joelle Pineau to pitch an “AI for All” framework. The choice of Cohere was a strategic flex for the Canadian ecosystem; the Toronto-based enterprise AI heavyweight reportedly scaled its annualized recurring revenue (ARR) to $240 million in 2025, proving that sovereign, enterprise-focused alternatives can commercially compete with Silicon Valley incumbents.

Public Capital Meets Practical Testbeds

To anchor these policy ambitions in commercial reality, the federal government’s economic development agency, PacifiCan, detailed a series of capital injections designed to move AI out of academic labs and into high-value infrastructure.

A fresh $1.8 million investment is targeting local AI testbeds, mapping out immediate enterprise use cases:

• Autonomous Mobility: Scaling pilot deployments at the Vancouver International Airport.

• Clinical Health Tech: Implementing automated pathology image analysis within the Provincial Health Services Authority.

This follows an existing $9.9 million federal allocation via Innovate BC, signaling a coordinated Canadian playbook to leverage public procurement as a growth engine for domestic AI vendors.

The Bottom Line

For founders, enterprise buyers, and venture capitalists, Web Summit Vancouver signaled that the AI landscape is shifting from an era of raw engineering breakthroughs to a complex game of compliance, jurisdiction, and IP defense.

As public institutions increasingly tie procurement dollars to localized data controls, the real winners of the next phase of the AI boom may not be the teams with the largest models, but those who best navigate the tightening knot of national borders and copyright law.