You’ve heard the mantra: prompt engineering is the must-have skill in today's AI-enabled world. But in conversations with Canadian firms at the front edge of the AI rollout, we’re hearing the real productivity unlock is knowing when to use AI, how to adopt it, and why it matters – what we call AI fluency.
If AI literacy is the ability to use the tools, AI fluency is the ability to go beyond: to identify an opportunity or problem, decide whether – and how – AI should address it, and run a pilot to prove it. For employers, these skills can spell the difference between simply staying afloat in the AI wave and converting it into real gains in productivity and innovation. So what is AI fluency, how is it different from AI literacy, what does it look like on the job, and how do we develop it in people?
Three ways AI fluency shows up on the job
1. The Opportunity Spotter
When “add AI” feels like the default move, the Opportunity Spotter asks a sharper question: where will AI create value, and where will it only add complexity? At D2L, a homegrown global software company focused on online teaching and learning, product leads who have layered AI into learning tools for years make tighter calls about when an AI feature truly improves learner and client support. They stay close to users, define the real problem, and run small pilots to test time and cost. They check data needs early and right-size the plan before it turns into a costly rework later.
You see the same discipline at Loblaw Digital, the team that designs, builds, and operates Loblaw’s digital experiences, where AI-powered personalization aims to increase customer satisfaction and basket size. Opportunity Spotters decide which features to keep and which to retire when real-world tests don’t meet the mark. They turn customer feedback into product changes, not just reports; choose whether to build, buy, or partner based on what works operationally; and can name the simplest real-world test that shows the feature delivers value.
The competencies: systems thinking; commercial and operational awareness; ROI assessment; risk analysis; customer responsiveness.
2. The Relationship Architect
As more interactions are mediated by AI, the differentiator is knowing when human connection is not replaceable. Relationship Architects map the user journey end-to-end, identifying moments where reassurance, empathy, clarity, cultural nuance, or decision confidence are non-negotiable, and design the handoff between digital agent and real human.
Passage, an international talent firm that uses AI to match skilled talent with labor shortages in Canada, puts this into practice with multilingual AI agents that guide international students through school admissions, loan applications, and job matching, then adds the human layer where it matters most. On the ground, that means mapping the journey, flagging the key moments that matter, setting simple guardrails to protect dignity and privacy, and tracking where students get stuck so teams can step in before trust slips. Their success rests on knowing what students value in human interactions, which is why they also hire student success advisors alongside the agents.
You see the same logic in global consulting firm McKinsey’s “hybrid intelligence” approach: they don’t sell AI in isolation; they sell AI fused with human judgment, ethics, and domain expertise. Relationship Architects make that fusion concrete, framing what clients want from a consulting relationship (confidence in recommendations, explainability, accountable ownership), selecting where AI accelerates insight, and specifying when a partner must be on a call. They write escalation rules, set consent and transparency standards, and ensure outputs are explainable to decision-makers.
The competencies: AI fluency applied to relationships; emotional intelligence; user-journey mapping; and communication tied to measurable outcomes (resolution time, completion rates, satisfaction).
3. The Anticipator
While most teams use AI to shave time off today’s tasks, Anticipators ask: what becomes possible that wasn’t before, and how do we prepare for it? In Canadian HR tech firm Knockri, AI assessments don’t just streamline hiring; they widen access to under-represented talent, open new growth areas for employers, and nudge policy toward fairer practices. The work is future-looking and practical: watch the trends, think through the knock-on effects, and run small tests that de-risk a bigger bet. On the job, that looks like turning a “bias reduction” goal into measurable targets, building feedback loops with hiring managers and candidates, and publishing explainable criteria so trust scales with adoption.
You see the same approach in the public sector. Across levels of government, urban planning teams are piloting machine learning to forecast housing, transit, and climate-resilience needs, shaping investments that prevent tomorrow’s gaps. Anticipators translate forecasts into siting choices, permit timelines, and infrastructure phasing the public can understand. They convene cross-functional partners early (data, legal, community), stress-test scenarios against equity and resilience goals, and design reversible steps so plans can adapt as conditions change.
The competencies: future vision; trend sensing; systems design; cross-functional fluency; disciplined problem-solving.
The Business + Higher Education Connection
Canada doesn’t just need AI users — we need AI deciders. That’s AI fluency: judging where AI creates value; managing risks; designing human-tech relationships that earn trust; and looking around the corner to build what comes next.
Where to start:
- Employers: map your workflows by time and cost; flag 2-3 AI use cases; run a 90-day AI pilot (with a clear owner and KPIs); define an AI fluency skills map per role (problem framing, ROI sizing, risk, data readiness); bake AI fluency into PD and org-wide upskilling/reskilling aligned with level + stage of career; add interview prompts and a practical task (mini business case or pilot plan) to hiring.
- Post-secondary institutions: launch faculty PD on AI adoption cases, skills, and assessment; embed AI fluency across computer science, business, arts/humanities, and continuing ed (short module or micro-credential); assess learners with a one-page business case + pilot plan + risk/ethics note (marked with a shared rubric).
- Together (via work-integrated learning): co-create WIL where student teams deliver a scoped AI pilot for an employer (baseline, KPI, data plan, ethics note); use a common scorecard (value, risk, usability, change plan); hold a short debrief to decide whether to keep it, change it, or stop it; offer micro-credentials co-badged by employers when learners demonstrate AI fluency.
If employers and post-secondary institutions train for AI fluency, we’ll build deciders, not just users, and Canada’s productivity will follow.