How AI is redesigning jobs instead of replacing them, according to Indeed's latest labour report

Job vacancy rate in Canada edged down from 3.1% to 2.8% between Q3 2024 and Q3 2025, and unemployment finished 2025 at 6.5%. Indeed research finds roughly 25% of jobs are highly transformable by GenAI, while a large share face moderate or hybrid transformation with AI augmenting tasks rather than fully replacing workers. Employers are urged to prioritize job redesign and workforce development to integrate AI and protect retention, but the near-term macro and market implications are limited.

Analysis

Macro tailwinds (trade fragmentation, cautious immigration) are pushing corporate spend away from headcount growth toward productivity investments; that reallocation favors vendors that can convert fewer people into the same output via software, tooling and training rather than pure labor arbitrage. Expect a 6–24 month rotation of capex from terminals and offices into cloud AI features and learning platforms, with margins concentrated at vendors who own the integration layer between HR systems and AI models.

Second-order competitive dynamics will bifurcate the HR ecosystem: incumbents that bundle applicant tracking, payroll and embedded L&D will capture wallet share from point solution vendors, while regional staffing firms with deep local networks become the go-to for near-shore labor as trade risk drives firm-level regionalization. A corollary is that cloud/AI infra providers will monetize these flows through subscription + usage economics, making their enterprise relationships stickier and increasing revenue multiplicity over time.

Key risks are regulatory and demand-side: rapid tightening on GenAI data use or model outputs would compress near-term adoption (3–12 months), while a macro slowdown or large AI-related scandal could reverse corporate hiring strategies and leave software vendors with deferred implementation revenue. Conversely, persistent labor scarcity in core roles (2–5 years) would accelerate training spend and favor integrated HR/SaaS platforms, so monitor hiring efficiency metrics and enterprise AI commitments as leading indicators.

For portfolio construction, prioritize exposure to platforms that can upsell training and AI modules, hedge pure-play gig labor names and size AI-infra exposure using option structures to protect against a deep post-hype drawdown. Trade timing should target near-term reporting windows and major regulatory milestones, since both are likely to be binary catalysts for valuation reratings.

Key Decisions for Investors

• Long Workday (WDAY) 12–24 months: buy shares on pullbacks or buy 12–18 month call spreads to capture upside from cross-sell of learning/AI modules into an installed HR base. Risk/reward: asymmetric — ~30–50% upside if adoption accelerates vs downside limited to premium (or ~20% equity drawdown if implementation stalls).
• Pair trade: Long ManpowerGroup (MAN) 6–12 months / Short Upwork (UPWK) 6–12 months — idea is to capture a shift from gig marketplaces to local staffing and managed services as trade/regionalization increases. Size equal dollar exposure; target 20–25% gross return if staffing reacceleration outpaces gig demand, stop-loss at 12% adverse move.
• Tactical AI-infra hedge: Buy Nvidia (NVDA) 6–12 month call spreads (buy ATM, sell 20–30% OTM) to participate in continued AI compute demand while capping cost; reward if enterprise model deployments scale, protection if chip cycle mean-reverts. Expect 2:1 upside vs premium risk in base case.
• Long Microsoft (MSFT) 9–18 months on strength in enterprise AI bundles: buy shares or sell covered calls to generate yield; monitor monthly cloud AI usage metrics and major enterprise HR contract announcements as triggers for take-profit (target +15–25%) or add-on.
• Event hedge: Buy 3–6 month protection (puts) on high-multiple pure-play HR/point-solution names if regulatory noise around GenAI increases post any major data/privacy hearings — protects against binary downside from policy shocks that could delay deployments.