Key Takeaways From All In 2026: AI Sovereignty, Infrastructure And Canada’s Growing Role

This week, technology leaders, government officials and industry executives gathered in Montréal for ALL IN 2026, a leading forum at the interp of technology, investment and national strategy. The discussions focused on issues with direct implications for Canadian businesses: who will control the future of artificial intelligence, and what does that mean for companies operating here?

The panels showed Canada deliberately pursuing global AI-hub status, with implications for investment, regulation, data governance and competition.

A Canadian-German AI Merger

Cohere, a Canadian AI firm, signed a definitive merger agreement with German AI firm Aleph Alpha, creating a transatlantic AI company.

• What happened. Cohere CEO Aidan Gomez announced the merger as a “Canadian-German champion at the AI frontier” to reduce global reliance on Silicon Valley.
• Why it matters. Frontier AI requires billions in capital; the deal reflects an emerging “middle power” strategy in which Canada and Germany pool resources to compete.
• Government backing. Canadian Minister of Innovation Evan Solomon and German officials signalled strong public-sector support; discussions on Canadian government funding are active.
• Data sovereignty. AI-era sovereignty means preserving choice, protecting sensitive domestic data under local laws and avoiding forced reliance on foreign tech monopolists.

Canada And Germany Invest In Lawzero: A New Path To Safe AI

Canada and Germany committed up to CAD $300 million in joint funding for LawZero, a Montréal-based not-for-profit founded by Turing Award winner Yoshua Bengio, to develop a fundamentally new form of advanced AI designed to be trustworthy, safe and capable.

• What is LawZero. LawZero is a not-for-profit AI safety research organization incubated at Mila – Québec AI Institute. Its core research centres on using probabilistic reasoning to understand and predict the world without pursuing goals of its own.
• Scale of investment. Canada and Germany will each contribute up to CAD $150 million and EUR $100 million, respectively. The investment will fund hiring, dedicated sovereign computing infrastructure built in partnership with Canadian firms Hypertec and 5C and a new LawZero office in Germany.
• Why it matters for clients. This investment signals governments are moving beyond regulating AI after the fact and are now directly funding alternative technical approaches to AI safety. For clients in regulated industries, financial services, health care and critical infrastructure, this investment could accelerate the availability of AI systems that are “safe by design” and easier to adopt within existing compliance frameworks.
• A surprising shift. The fact that two G7 governments are willing to back this approach with public funds suggests a growing recognition that the current trajectory of frontier AI development carries risks that market forces alone will not address.
• What Canadian businesses should consider. Businesses should be aware that sovereign AI infrastructure and safety-first development are becoming explicit policy priorities for the Canadian government, which may influence future procurement standards and funding eligibility criteria.

A New Approach To AI Safety

Cohere’s Gomez proposed a practical alternative to the fear-driven AI-safety debate: a structured framework shifting the burden of proof to developers.

• The critique. Gomez said some industry players use extreme “extinction risk” narratives to justify anti-competitive regulation.
• A three-pillar framework. Gomez proposed: (i) defining risks on precise, measurable axes; (ii) requiring developers to prove their technology safe before deployment; and (iii) public incident reporting when models fail.
• Who sets the rules. Government-led, multi-stakeholder processes—not a small group of technology executives—should set safety standards.

Bell’s Bet On Canadian AI Infrastructure

Bell outlined one of the largest domestic technology infrastructure projects in Canadian history: its Saskatchewan-based “Bell AI Fabric” initiative.

• Scale of investment. Total initiative capital could exceed $50 billion; Bell is committing approximately $5 billion directly to data centre facilities.
• Canadian Sovereignty Island. Bell has partnered with Canadian firms, including Cohere and Hypertec, to deliver end-to-end AI solutions entirely within a secure, Canadian-controlled jurisdiction.
• Environmental design. Saskatchewan data centres use closed-loop cooling with zero industrial-water consumption; waste heat will power a local district energy grid for the George Gordon First Nation.
• Community and economic impact. More than 500 construction workers are on-site; Bell is training local First Nation residents for permanent roles, and service credits are being offered to Saskatchewan colleges and universities.

OpenAI Deepens Its Canadian Presence

OpenAI identified Canada as a priority market, announcing new partnerships and signalling potential infrastructure investment.

• Countries Program. OpenAI’s government-facing initiative works with national administrations to guide safe AI adoption, treating AI as a general-purpose technology like electricity.
• Why Canada. OpenAI cited Canada’s political stability, clean-energy grid and progressive regulatory environment as attractions for partnerships and potential data centre development.
• Cybersecurity. OpenAI recently trained 40 Canadian critical-infrastructure stakeholders in AI-powered cyber defence; Canada was among the first countries to access its latest defensive systems.
• New partnerships. OpenAI announced partnerships with an Ontario-based media firm and Creative Destruction Lab, a Toronto-based startup incubator.
• Regulatory vision. OpenAI endorsed a “Technology Stability Board,” modelled on the Financial Stability Board, to coordinate international AI safety standards.

What “Technology Sovereignty” Means For Business

A closing panel examined technology sovereignty in practice and its implications for organizations managing data, infrastructure and vendor relationships.

• Data residency is not enough. Storing data in Canada does not guarantee sovereignty if the managing corporate entity is subject to foreign extraterritorial laws; true security requires local operational control of both hardware and governance.
• Risks beyond AI. Sovereignty vulnerabilities also include physical infrastructure—subsea cables, satellite constellations and cross-border digital identity standards.
• Government as anchor tenant. Governments should purchase domestic data-centre capacity for critical public workloads, providing steady revenue to de-risk private expansion.
• Pension fund capital. Canada’s world-class pension funds are a powerful vehicle for long-term investment in sovereign digital infrastructure.
• Collaborative approach. Middle powers should pool resources on joint open-source or open-weight AI projects to gain strategic leverage in a more competitive geopolitical environment.

Conclusion

ALL IN 2026 showed Canada positioning itself as a serious contender in the global AI race through strategic mergers, massive infrastructure investment and sovereignty-first data and technology governance.

For Canadian businesses, the rules governing AI investment, data residency and vendor relationships are shifting rapidly; organizations that fail to adapt risk being left behind.