AI-ready factories will still need people: Manufacturing’s future depends on skills and fibre

Manufacturing is entering a new phase of digital transformation. The next generation of factories will not simply be automated; they will be connected, data-rich, AI-enabled and increasingly dependent upon resilient digital infrastructure. Yet one of the most important findings from Deloitte’s 2026 Manufacturing Industry Outlook is that people remain central to this transformation. More than 81 percent of manufacturing task hours are expected to remain human-driven, even as manufacturers increase their adoption of advanced automation, agentic AI and physical AI systems. Deloitte also reports that planned use of physical AI is expected to rise from 9 percent today to 22 percent within two years, indicating a rapid shift towards more autonomous industrial systems.

This creates an important paradox for manufacturers. The factory of the future will be more technologically sophisticated, but its competitive advantage will not come from technology alone. It will come from the ability of the workforce to operate across engineering, automation, AI, data and connected manufacturing systems. The people who succeed in advanced manufacturing environments will increasingly be those who can interpret data, understand production processes, work with digital twins, interact with intelligent machines, and make risk-based decisions when automated systems require human oversight.

Deloitte’s outlook notes that manufacturers are investing in automation hardware, data analytics, sensors and cloud computing as foundational tools for smart manufacturing, while also identifying talent and skill development as key constraints.

For companies building digital infrastructure technologies, including fibre networks, AI data centres, industrial automation platforms and connected manufacturing systems, this creates a new type of skills demand. Fibre connectivity is no longer a background utility. It is becoming a core component of industrial competitiveness. Low-latency, reliable connectivity allows production equipment, sensors, AI models, manufacturing execution systems and enterprise platforms to communicate in near real time.

Canada’s AI strategy

In an AI-ready factory, connectivity becomes as important as power, water, compressed air or clean utilities. Without robust fibre and network architecture, manufacturers will struggle to deploy predictive maintenance, real-time quality monitoring, autonomous logistics or AI-supported production scheduling at scale. Canada’s own AI strategy recognises that AI infrastructure includes not only models and applications, but also data centres, cloud, cooling, connectivity, networking, chips, servers and national telecommunications networks.

Canada provides a particularly interesting case study because it is trying to align AI capability, sovereign compute, telecommunications infrastructure and industrial competitiveness. The federal government’s national AI strategy identifies manufacturing and robotics as priority sectors, noting that persistent labour shortages and reshoring pressures make industrial AI and robotics essential to advanced manufacturing and defence production. The same strategy highlights Canada’s AI ecosystem, including more than 150,000 AI-related jobs and over 3,500 Canadian firms developing advanced AI models, tools and applications.

However, the Canadian challenge is not only about inventing AI technologies. It is about adoption. Canada’s strategy acknowledges an adoption gap, with only 12 percent of Canadian businesses using AI to produce goods or services between mid-2024 and mid-2025, rising to 14.5 percent planning to do so by mid-2026. Among small and medium-sized enterprises, AI adoption was around 8 percent. This means that Canada has strong research capacity and digital ambition, but many manufacturers still need support to convert AI potential into operational capability.

The workforce implications are significant. A report from NGen on artificial intelligence in Canadian manufacturing argues that AI can improve operational efficiency, product quality, predictive maintenance, robotics, computer vision and supply chain optimisation. At the same time, the report stresses that AI integration requires investment in workforce upskilling, including data analytics, computer science, programming, robotics and digital literacy. The report’s case studies suggest that AI often reorganises work rather than simply eliminating jobs, shifting employees towards higher-value tasks where human judgement remains important.

The factory floor is becoming a connected system, and the workforce must be organised accordingly.

This is where manufacturers need to rethink training. Traditional training models, where engineers, production operators, IT specialists and quality teams are taught in separate functional streams, are increasingly inadequate. AI-ready manufacturing requires multidisciplinary learning. Operators need to understand data integrity and sensor outputs. Engineers need to understand AI model limitations and automation governance. IT teams need to understand production risk, uptime constraints and quality requirements. Quality professionals need to understand how algorithms, data pipelines and automated decision systems affect product quality, batch release, traceability and regulatory compliance.

The same applies to fibre and digital infrastructure companies. Their future success will depend on teams that combine fibre engineering, automation, data centre operations, quality systems, cyber resilience, digital manufacturing and data-driven operations. This is no longer simply a telecommunications or network-engineering challenge. It is an industrial capability challenge. A network that supports an AI factory must be designed for resilience, security, latency, scalability and operational continuity. These features become essential when manufacturing systems depend on real-time data flows.

Canadian public private partnerships

Recent Canadian investments show the direction of travel. Bell Canada and the Government of Saskatchewan announced a 300 MW AI data centre in the Rural Municipality of Sherwood, near Regina, with the facility linked to Bell’s national fibre backbone through a partnership with SaskTel. Bell describes the development as a major expansion of domestic AI compute capacity, with a significant portion dedicated to sovereign AI compute so that Canadian government agencies, researchers and enterprises can access AI infrastructure while keeping data within Canada.

TELUS has also announced plans to invest more than C$66 billion through 2030 to expand and enhance network infrastructure and operations across Canada. The company points to its PureFibre and 5G networks as critical infrastructure for productivity, innovation and competitiveness, while also linking connectivity investments to Canadian AI leadership and sovereign AI infrastructure. TELUS states that its 5G network reaches more than 90 percent of Canada’s population, while its PureFibre network reaches 3.7 million households and businesses.

These investments matter for manufacturing because AI adoption depends upon infrastructure that can support heavy data movement, continuous monitoring and secure industrial control. A smart manufacturing system cannot function effectively if the underlying network is fragmented, unreliable or insecure. Predictive analytics, computer vision inspection, autonomous vehicles, cobots, digital twins and AI-based scheduling all require dependable data transmission. As the NGen report observes, AI applications in manufacturing include robotics, machine learning, predictive analytics, computer vision, supply chain optimisation, design and testing, and efficiency improvement. These use cases are only as strong as the digital infrastructure supporting them.

The business opportunity is therefore broader than installing robots or deploying AI software. It involves building an operating model where human expertise, intelligent automation and connected infrastructure work together. Manufacturers that view AI as a labour replacement tool may miss the larger opportunity. Deloitte’s outlook is clear that human capabilities such as creativity, collaboration, critical thinking, adaptability and emotional intelligence remain essential. It further suggests that AI can accelerate training, knowledge-sharing and remote collaboration, helping manufacturers address long-standing talent challenges.

For manufacturers, three priorities stand out. First, companies need to develop multidisciplinary talent with expertise across engineering, AI, automation, digital systems and data. Second, they need to embed continuous learning so employees can adapt as intelligent manufacturing technologies evolve. Third, they need to break down traditional silos between production, IT, engineering, digital transformation and quality teams.