The AI Paradox
Nearly every Canadian tech company is using AI—but are they doing it right?
Our October 2025 Member Pulse Survey revealed a striking paradox: 95% of tech companies are using AI in their business processes, yet only 55% have a strategy for it. Even more telling, organizations are split almost evenly on whether it’s ethical to use AI in talent decisions (35% agree, 32% disagree).
This disconnect points to a bigger issue: we’re adopting AI tools without thinking through the implications.
As Rocky Ozaki, Founder and CEO of The NoW of Work, put it bluntly at our recent Tech Talent North conference: “Don’t impose AI on teams. You’ll get an immune reaction.”
Why AI Adoption Is Broken
The problem isn’t the technology—it’s how we’re implementing it. Organizations are rushing to “do AI” without asking the fundamental question: what problem are we actually solving?
Here are some common pitfalls we’re hearing about:
- Tools over problems – Implementing AI for the sake of “doing AI” rather than addressing real user pain points
- Lack of governance – No clear policies or ethical guidelines (only 50% of surveyed organizations have formal AI policies)
- Unknown readiness – Organizations don’t know if their teams are prepared to use AI effectively
- Vanity metrics – Measuring adoption rates instead of actual impact (only 1 organization in our survey tracks metrics for AI gains)
- Bad data – AI trained on poor quality data won’t solve problems, it will create new ones
The result? AI tools that nobody uses, or worse, tools that introduce bias and erode trust.
What the Data Tells Us
Our survey of 44 Canadian tech companies (ranging from 25 to over 1,000 employees) shows AI is being used most commonly for:
Top Uses Across Organizations:
- Written work using tools like ChatGPT (93%)
- Condensing/summarizing research (78%)
- Idea generation/brainstorming (73%)
- Creating visuals (59%)
- Rote tasks for efficiency (54%)
In HR and People & Culture specifically:
- Responding to standard policy questions (27%)
- Screening candidates (20%)
- Responding to employee relations issues (20%)
These numbers in talent and workforce management are climbing—they’ve doubled in the past 18 months—but they’re still relatively low. We’re clearly in the early adopter stage for structured AI use in HR.
The Productivity Promise (and the Anxiety)
The good news: 64% of organizations say AI has already enhanced HR/P&C team productivity. Looking ahead three years, 73% believe it will improve productivity further, and 59% think the quality and output of work will improve.
The concerning news: 46% think less headcount will be required to do the same amount of P&C work.
This anxiety is real, and it’s exactly why the human-centered approach matters. As Kaitlin Graves, VP People at Pagefreezer, observed after Tech Talent North: “Tools don’t transform companies, people do. All the AI in the world won’t help achieve business goals if your teams don’t trust it or feel empowered to use it.”
A Better Way: Human-Centered AI
What does human-centered AI actually look like? Here’s a framework built from Rocky Ozaki’s Tech Talent North closing keynote, The Hidden Superpower of P&C: Leading the AI Agenda:
1. Bottom-Up, Not Top-Down
Like living core values, AI adoption works best when it comes from the ground up. Make it opt-in, not imposed. Democratize the journey rather than mandating tools from above.
2. Solve People Problems First
Know what problem you’re trying to solve before implementing any AI tool. If you don’t solve your users’ pain points, you’ll have expensive technology that sits unused. AI adoption happens when you make people’s lives easier.
3. Augment, Don’t Just Automate
Focus on how AI can enhance human capabilities, not just replace them. This is the difference between tools that empower teams and tools that create fear.
4. Build in Bias Awareness
Ask your AI vendors what testing they do to minimize bias, and how frequently. Make this a non-negotiable part of your procurement process.
5. Keep Humans in the Loop
AI should inform decisions, not make them autonomously—especially in areas like talent management where human judgment is critical.
6. Have a Strategy for Your People
Human-centered AI requires clear plans for:
- Upskilling – Building AI literacy across your organization (in April 2024, only 18% of People & Culture teams rated themselves as familiar or very familiar with AI; by October 2025, that jumped to 47%)
- Reskilling – Preparing people for evolving roles
- Redeployment – Thoughtfully transitioning people as work changes
Learn more about Rocky’s work here.
The Critical Thinking Imperative
Here’s something that should concern us all: according to a Semrush study, Reddit outranks industry experts in professional AI search citations. As Mohamed Yousuf, CEO and Co-Founder of Smart Workforce AI, points out: “40% of what AI tells you comes from Reddit. Not research papers or expert databases. Reddit threads.”
We’re hearing that AI usage is reducing our critical thinking skills. But if we’re using it right, we should be exercising those skills more—by fact-checking, comparing sources, questioning outputs, reasoning through recommendations, and forming our own conclusions.
This is why the human element isn’t just nice to have—it’s essential.
The Questions You Should Be Asking
If your organization is among the 95% using AI, ask yourself:
- Do we have a clear strategy and policy for AI use? If not, what’s stopping us?
- Are we implementing AI to solve real problems, or for vanity? Can we articulate the specific pain points we’re addressing?
- Have we discussed the social and ethical implications? (Only 62% of surveyed organizations have had this conversation at the senior leadership level)
- Do our teams trust and feel empowered to use AI tools? Are we getting an “immune reaction”?
- Are we measuring what matters? Beyond adoption rates, how are we tracking actual impact?
- How are we preparing our people for an AI-augmented future?
Moving Forward
AI adoption is less about technology and more about change management and transformation. The organizations that will succeed aren’t necessarily the ones with the most sophisticated tools—they’re the ones that put people at the center of their AI strategy.
Start with the problems you need to solve. Invest in clean data. Build governance frameworks. Prioritize upskilling. Keep humans in the loop. And above all, remember: the goal isn’t to do AI for AI’s sake. It’s to make work better for the people doing it.
TAP Network members: We’ll continue exploring AI implementation best practices in upcoming events and roundtables. Learn more about membership benefits →
Data in this article comes from TAP Network’s October 2025 Member Pulse Survey of 44 Canadian tech companies, ranging from 25 to over 1,000 employees, as well as insights from Tech Talent North 2025.
