Insurers Increase Their Bets on AI-Focused Stocks Amid Surging Valuations
Insurers spent the back half of 2025 as steady net buyers of AI-linked equities, a sharp turn from 2023 and 2024. Fresh data from Clearwater Analytics shows consistent additions to Nvidia, Microsoft, Alphabet, and Meta, measured using 90-day moving averages to filter out noise. That methodology points to intent, not impulse. It also signals a broader shift in how insurance balance sheets are leaning into growth.
What changed in 2025
AI leaders kept pulling away. Valuations sprinted higher as semiconductor performance, cloud demand, and generative AI adoption accelerated across enterprise workflows. Nvidia's infrastructure role, and the hyperscalers' cloud and model stacks, stayed central to the thesis.
At the same time, critics raised fair flags: elevated multiples, aggressive multi-year revenue assumptions, and funding loops across private markets. The question isn't whether AI matters-it's whether the current pricing fully bakes in execution risk and competitive pressure.
Why insurers are buying now
Two forces are at work. First, fixed income did its job, but yield volatility and reinvestment risk kept CIOs looking for a second engine of long-term return. Second, technology concentration has been where growth lives, and the clearest runway sits in AI infrastructure, software, and platforms.
Clearwater's flow analysis, built on longer lookbacks, suggests insurers are increasingly comfortable with equity volatility if the thesis is durable. For many carriers, the move is less about chasing momentum and more about matching liabilities with assets that can participate in long-cycle productivity gains.
The bull and bear cases, in plain terms
Bull case: Demand looks real. Enterprises are spending on data centers, cloud services, and AI software tied to measurable outcomes-automation, analytics, and better customer experiences. Policy tailwinds and ecosystem maturity support multi-year investment.
Bear case: Pricing leaves little margin for error. Competition is heating up, supply chains can bottleneck, and any slowdown in capex or model monetization can compress multiples fast. If expectations stay sky-high, even "good" quarters can disappoint.
Portfolio implications for carriers
- Define exposure bands: Set clear guardrails by sub-theme (semis, cloud, models, software) and by single-name concentration. Make the limits explicit and revisit quarterly.
- Stage entries: Use staged allocations and time-weighted buys to reduce timing risk. Moving-average triggers or calendar ladders help avoid reacting to headlines.
- Stress test the thesis: Run scenarios on multiple compression (e.g., 20-40%), capex pauses from hyperscalers, and margin pressure in AI software. Tie each to portfolio P&L and capital ratios.
- Mind liquidity and capital: Check RBC and accounting treatment, especially for large-cap tech held in the equity sleeve. Confirm daily liquidity for potential rebalance needs.
- Balance factor risk: AI leaders are also mega-cap growth. Know your overlap with broad indices and consider how that interacts with your SAA targets.
- Governance cadence: Set a standing review for AI exposure, with pre-agreed actions if valuations or fundamentals break thresholds.
What to watch next
- From capex to revenue: Do GPU and data center builds translate into sustained software spend and customer adoption?
- Unit economics: Evidence that AI features boost retention, pricing, or margins-beyond pilots and proofs of concept.
- Supply and energy constraints: GPU availability, networking gear lead times, and power capacity for new data centers.
- Competitive dynamics: Model quality gaps, open-source traction, and pricing shifts across cloud providers.
- Policy and compliance: Clear rules can reduce risk premiums; uncertainty can do the opposite.
How insurers can turn thesis into process
- Map exposures to use cases: Connect each holding to a specific cash-flow driver (AI infrastructure, cloud attach, enterprise software upsell). If the driver is unclear, trim or exit.
- Build a signals dashboard: Track hyperscaler capex, major model releases, AI workload utilization, and pricing for inference and training. Tie signals to action thresholds.
- Use risk budgets: Allocate a defined risk budget to AI and require offsets elsewhere when exceeded.
- Consider hedges prudently: Where policy allows, evaluate low-cost index or factor hedges to blunt drawdowns without dismantling core positions.
Bottom line
Insurers shifting from net sellers to net buyers of AI leaders looks less like a fad and more like a structural reweight. The bet is simple: mission-critical infrastructure and software will keep earning wallet share, even if quarters wobble.
Discipline still decides outcomes. Keep entries paced, size positions with intent, and test assumptions against hard data. If the leaders turn innovation into scalable, profitable growth, today's allocations can compound without constant heroics.
Source and further learning
For data on insurer flows and holdings analysis, see Clearwater Analytics.
If your team is building AI literacy for investment and product lines, explore AI courses by job to accelerate practical skills across the organization.
Your membership also unlocks:
