AI Margin Gains Outside Tech May Take Years — What It Means for Moats, Spending and Jobs

Apollo chief economist Torsten Slok warned in a May 2026 note to clients that AI-driven margin gains outside regulated industries such as healthcare, banking, and pharma could take "well beyond" what Wall Street expects, noting that process overhauls and privacy rules may delay productivity boosts by years rather than months.

AI-Driven Margin Gains Outside Tech May Be Delayed for Years — What It Means for Non‑Tech Competitive Moats

Torsten Slok argues that in heavily regulated sectors, the integration of AI requires lengthy process overhauls to satisfy compliance and privacy standards, which can stretch implementation timelines from months to several years. (Analyst view — Apollo) This delay means that firms outside tech may not see the margin expansion that investors have priced into their stocks.

Without near‑term AI‑driven cost savings, competitive moats in industries like banking or pharmaceuticals are unlikely to widen through operational efficiency gains alone. Companies may continue to rely on legacy cost structures while tech peers use AI to lower expenses and increase pricing power. (Analyst view — Apollo)

Consequently, the relative advantage of technology firms could persist longer than markets anticipate, reinforcing the sector’s valuation premium. Investors should reassess the durability of moats in non‑tech holdings when modeling AI impact.

Delayed Productivity Boosts Could Lead to Overinvestment in AI Infrastructure — Implications for Capex Cycles

Slok’s warning that AI‑enabled productivity gains outside tech may take five years instead of five months suggests that the current wave of AI capital expenditure could outpace actual earnings contributions for many firms. (Analyst view — Apollo) If businesses delay deployment, the demand for AI chips, data‑center capacity and related services may soften sooner than expected.

Infrastructure vendors such as semiconductor makers and cloud providers could experience a slowdown in order growth as non‑tech customers postpone purchases. This mismatch between supply‑side investment and demand‑side adoption creates a risk of excess capacity in the AI supply chain. (Analyst view — Apollo)

For investors, the implication is a potential reassessment of growth forecasts for AI‑related capital goods companies, with a closer look at order backlogs and customer concentration in regulated sectors.

Slower AI‑Enabled Job Displacement in Regulated Sectors — How Labor Markets May React

Because AI‑driven process changes in healthcare, banking and pharma are expected to be delayed, the pace of job displacement linked to automation may also be slower than projected. Workers in these industries could see‑retrain before large‑scale role reductions occur. (Analyst view — Apollo)

This slower transition may give policymakers and firms more time to design reskilling programs, potentially mitigating the social disruption often associated with rapid AI adoption. However, it also means that productivity‑related wage pressures could remain subdued for longer.

Investors watching labor‑intensive stocks should note that any near‑term earnings uplift from AI‑driven headcount reductions is unlikely to materialize quickly, keeping labor cost forecasts more stable in the short to medium term.

Tech‑Sector AI Stocks Face Repricing Risk if Expectations Are Not Met — Valuation Outlook

Slok notes that if AI profit gains outside tech take years rather than months, many AI‑themed stocks could experience a painful repricing as growth expectations are adjusted downward. (Analyst view — Apollo) The market has priced in rapid margin expansion across a broad range of industries, not just software and semiconductors.

When the anticipated earnings boost fails to appear on schedule, valuation multiples for AI‑linked equities may compress, especially for companies whose valuations rely heavily on long‑term AI‑driven growth assumptions. This could trigger a sector‑wide rotation toward firms with more immediate, demonstrable AI monetization.

Investors should therefore scrutinize the earnings guidance of AI‑exposed companies for concrete, near‑term margin impacts rather than relying on multi‑year projections.

Regulatory Hurdles and Privacy Rules as the Primary Bottleneck — Why Process Overhauls Take Time

The core reason for the delay, according to Slok, is that regulated industries must navigate complex compliance frameworks and privacy legislation before AI tools can be deployed at scale. These requirements necessitate extensive process redesign, validation, and staff training, which inherently consume years rather than months. (Analyst view — Apollo)

Unlike the relatively permissive environment for AI experimentation in many tech firms, banks, insurers and drug makers face stringent oversight from entities such as the Federal Reserve, the FDA and European data‑protection authorities. Each change to a workflow may trigger a new review cycle, extending timelines.

As a result, the speed at which AI can transform operating margins in these sectors is gated by regulatory readiness, not just technological capability. Investors should factor in the pace of rulemaking and enforcement when estimating the AI impact on non‑tech earnings.

Key Developments to Watch

• Apollo Global Management earnings call (Q3 2026) — management’s commentary on AI‑related margin expectations will test the economist’s thesis.
• FDA guidance on AI‑enabled medical devices (by November 2026) — any easing or tightening of approval pathways will directly affect deployment timelines in healthcare.
• European Union AI Act enforcement date (by November 2026) — the rollout of compliance requirements will shape how quickly banks and insurers can adopt AI systems.

If AI‑driven profit gains outside tech truly take years rather than months, how should investors adjust their expectations for sector rotation and valuation multiples over the next 24 months?

• Margin expansion — the increase in a company’s profit as a percentage of revenue, often driven by lower costs or higher prices.
• Process overhaul — a comprehensive redesign of business workflows to incorporate new technologies while meeting regulatory requirements.
• Productive AI — artificial intelligence applications that directly improve output or efficiency, such as automating routine tasks or optimizing resource allocation.