State Street Says Any Tech Weakness Is Chance to Rebuild

State Street Equity Research Head Marija Veitmane remains constructive on technology stocks, citing the sector's strong earnings and profitability. She argues that potential regulation curbing development of frontier AI models could favor larger technology companies, as regulation is likely to be more supportive of established players.

Analysis

The investable implication is less a broad technology beta signal than a potential regulatory moat for scaled platforms. Compliance, model-governance, auditability and legal-cost burdens favor firms with proprietary data, cloud distribution and balance-sheet capacity—principally MSFT, GOOGL, AMZN and META—while raising the fixed-cost hurdle for smaller model developers and enterprise software vendors without differentiated AI monetization. This could widen the valuation and earnings-durability gap over the next 6-18 months, but much of that outcome is already reflected in mega-cap concentration and premium multiples.

Near term, rhetoric around slowing frontier-model development is unlikely to impair hyperscaler earnings unless it translates into binding restrictions on deployment, data access, or power procurement. The more immediate earnings sensitivity remains AI infrastructure utilization: MSFT/AWS/GCP need sustained cloud and inference demand to convert elevated capex into accelerating margins. Watch quarterly capex guidance, cloud growth and disclosed AI revenue contribution over the next 1-3 months; a capex step-up without corresponding backlog or cloud reacceleration would challenge the “regulatory winner” narrative and pressure AI-linked multiples.

The contrarian view is that regulation may be supportive to incumbents but still economically negative for the sector if it slows enterprise adoption or creates liability around high-value use cases. In that outcome, expensive infrastructure beneficiaries—NVDA and AVGO—face the greatest de-rating risk because their expectations embed rapid deployment, while software incumbents retain more diversified earnings bases. STT has no clear direct earnings read-through; this is institutional-market commentary rather than a company-specific catalyst, so it does not justify a standalone position in STT.

Key Decisions for Investors

• Maintain a 6-12 month quality-AI basket overweight in MSFT, GOOGL and AMZN versus an equal-weight technology benchmark; prefer platform owners with cloud distribution and compliance capacity over unprofitable AI application names. Reassess if combined cloud growth fails to stabilize/reaccelerate while capex guidance rises.
• Express relative-value caution via long MSFT / short a basket of high-multiple, lower-scale AI software exposure (IGV as a liquid proxy) over 1-3 months. The thesis is regulatory fixed-cost absorption and stronger AI monetization at MSFT; exit if enterprise software bookings materially accelerate or MSFT Azure growth disappoints.
• Do not add directional exposure to NVDA solely on regulatory-moat logic. Treat hyperscaler capex commentary and evidence of inference monetization as required confirmation; a deployment restriction that reduces compute demand would create asymmetric multiple risk despite near-term order visibility.
• Monitor EU AI Act implementation, U.S. federal agency rulemaking, and any restrictions on model release or training-data use. Rules focused on reporting, safety testing and procurement favor incumbents; rules imposing deployment liability or compute limits would invalidate the constructive sector read-through.