en.Wedoany.com Reported - Ottawa, April 28, 2026 local time — Canadian telecommunications company TELUS and Canadian enterprise accelerator L-SPARK have jointly launched the TELUS Sovereign AI Accelerator. This is a six-month specialized program targeting high-potential startups and scale-ups, providing direct access to Canada's fastest and most powerful TELUS Sovereign AI Factory, headquartered in Ottawa, Canada.
The first five companies selected for the accelerator span five sectors: retail, healthcare, robotics, enterprise software, and industrial automation, representing different technology directions at various commercialization stages within Canada's current AI startup ecosystem. Airy3D leverages its DepthIQ IP technology to simultaneously output 2D images and 3D depth maps from a single passive image sensor, providing low-power 3D perception solutions for robotics, automotive, and industrial automation scenarios. Codalio builds an AI-driven product and application development platform, helping startups deliver enterprise-grade applications faster. Edge Signal introduces physical AI into retail operations, enhancing revenue and customer experience through store-level intelligence. PataBid focuses on the vertical construction sector, using AI enterprise bidding software to cover full-process standardization and risk management for complex specialized trades. TopoLift creates a customizable intelligence layer that learns a client's business structure, outputting industrial-grade decision recommendations with contextual understanding. The selected companies also receive compute credits for the TELUS Sovereign AI Factory and one-on-one business mentorship from L-SPARK executive advisors, covering everything from product roadmap optimization to investor network building.
Compute infrastructure is the core pillar of this accelerator. The TELUS Sovereign AI Factory is located in Rimouski, Quebec, Canada. It is wholly owned, built, and operated by TELUS, equipped with NVIDIA H200 GPUs and NVIDIA Quantum 2 InfiniBand networking, supplemented by WEKA high-speed NVMe storage to eliminate I/O bottlenecks in large-scale training. This facility has been confirmed in the TOP500 rankings as the most powerful supercomputer in Canada. The factory operates using 99% renewable energy, features an advanced 2N cooling architecture, and can utilize free cooling 98% of the time annually. TELUS Chief Information Officer Hesham Fahmy pointed out that the challenge facing Canadian AI talent and founders today is not a lack of ideas, but a structural shortage of sovereign compute power; this collaboration with L-SPARK effectively opens enterprise-grade computing infrastructure, previously only accessible to large corporations, to startups, allowing these companies to train, fine-tune, and deploy AI models within a Canadian-controlled jurisdiction.
The supply-demand gap in sovereign compute is accelerating the pace of investment in Canadian AI infrastructure. The Canadian federal government proposed in its 2025 budget to invest approximately CAD 1.5 billion over five years in AI computing infrastructure. The TELUS Sovereign AI Factory officially began operations in September 2025, making it Canada's first fully sovereign AI computing facility. Canada's national AI strategy started early with deep research accumulation; Montreal, Toronto, and Edmonton form one of the world's highest-density clusters of AI talent, but the country has long relied on foreign public cloud platforms for scalable GPU infrastructure. This accelerator has set clear industry priorities—regulated and mission-critical sectors such as the public sector, healthcare, finance, critical infrastructure, and utilities are listed as the top priority, as these fields have strict requirements for data localization, audit trails, and Canadian legal jurisdiction.
As Canada's leading enterprise accelerator, L-SPARK has supported over 130 Canadian companies over the past decade, helping them raise over $200 million in follow-on funding. Leo Lax, Executive Managing Director of L-SPARK, stated that Canadian entrepreneurs are highly ambitious on their AI roadmaps, but the compute bottleneck is stifling execution speed. This partnership allows entrepreneurs, for the first time, to initiate complete product iterations with scalable, compliant GPU resources without needing to send data abroad or build their own infrastructure.
TELUS plans to further expand AI compute access for Canadian startups in the coming months through partnerships with more accelerators, incubators, research institutions, and innovation centers. The first five selected companies will retain full control over their data and intellectual property, with all model training and inference conducted entirely within Canada.
This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com
