Amazon, Palantir and 12 more top tech stock picks from UBS analysts

UBS said the AI data-center spending cycle remains in its early stages and identified 14 high-conviction technology, media and telecommunications stock picks, including Amazon and Palantir. The analysts see opportunities across the broader AI ecosystem as data-center investment and related spending continue to expand. The report is supportive for AI-exposed equities, though it provides no new company-specific financial estimates or target-price changes in the supplied text.

Analysis

The investable question is no longer whether AI capex rises, but whether spending broadens from hyperscaler-owned infrastructure into software monetization and enterprise deployment before valuation expectations outrun earnings. AMZN has a more balanced exposure than pure AI infrastructure: AWS can monetize both training and inference, while retail/logistics may ultimately capture productivity gains. The near-term earnings sensitivity remains cloud growth and AWS operating-margin conversion, not generic AI enthusiasm.

PLTR is the higher-beta expression of enterprise AI adoption, but its valuation leaves little tolerance for deceleration in commercial customer growth, net-dollar retention, or incremental deal size. A shift from pilots to production contracts should be visible within the next 1-3 quarters through remaining-deal-value growth and sustained operating leverage; absent that evidence, AI narrative multiples are vulnerable to a sharp compression even if revenue continues growing.

Consensus may be underestimating an AI-capex digestion phase: hyperscalers can sustain investment while vendors and enterprise customers face deployment bottlenecks from data quality, integration labor, power availability, and uncertain ROI. This favors platforms with distribution and customer data over companies selling undifferentiated model access. Over 6-18 months, inference economics and cloud price competition could shift value from GPU supply-chain beneficiaries toward AWS, Microsoft (MSFT), Alphabet (GOOGL), ServiceNow (NOW), and cybersecurity/data-governance vendors rather than speculative application names.

Key Decisions for Investors

• Prefer long AMZN versus short an AI-software basket proxy (IGV) over the next 3-6 months: AWS growth reacceleration plus retail-margin upside offers a more defensible earnings bridge than high-multiple software. Reassess if AWS growth fails to improve sequentially or AWS margin retreats despite capex scaling.
• Treat PLTR as a tactical, not core, long into the next earnings cycle only if commercial revenue growth and deal-value disclosures confirm production deployment. Use defined-risk call spreads rather than outright exposure; thesis is invalidated by material commercial-growth deceleration or weaker forward revenue guidance.
• Build a 6-12 month quality AI pair: long AMZN or MSFT / short a diversified unprofitable software ETF such as ARK Innovation (ARKK), sized modestly. The trade expresses monetization and balance-sheet durability against a potential multiple reset in AI-adjacent equities; stop if long-leg cloud growth materially underperforms software-sector growth.
• Monitor hyperscaler capex guidance, power-procurement commentary, and enterprise IT-budget surveys over the next 1-3 months. Evidence of capex restraint or delayed AI project conversion is a signal to reduce broad AI beta rather than add exposure.