3 Top Stocks to Buy in May

The article is broadly bullish on Taiwan Semiconductor, Amazon, and Lemonade, citing strong AI-related growth and improving fundamentals across all three. TSMC reported Q1 revenue up 41% year over year with gross margin expanding to 66.2% and guided Q2 revenue to +35%; Amazon posted accelerated AWS growth, triple-digit AI revenue growth, and 24% ad sales growth; Lemonade’s in-force premium rose 32% year over year with management guiding for positive adjusted EBITDA by year-end. The piece is opinionated stock-picking commentary rather than hard market news, so likely price impact is limited to the individual names.

Analysis

The common thread is not “AI beta,” but capex monetization power. TSM is the cleanest toll collector: hyperscaler spending is shifting from optional experimentation to multi-year infrastructure commitments, which tends to compress customer cadence risk and improve fabrication utilization, yields, and pricing leverage. The second-order effect is that every incremental AI buildout makes advanced-node scarcity more durable, which should keep the ecosystem disciplined even if headline revenue growth normalizes.

AMZN is the higher-quality operating leverage story because AI is not just a standalone revenue line; it is pulling through core cloud consumption, ad targeting, and fulfillment efficiency simultaneously. That creates a flywheel where AI adoption raises wallet share, not just gross profit per unit. The underappreciated risk is that markets may be capitalizing AWS acceleration as if it is linear, when the more likely path is lumpy but still durable; that favors buying weakness on any AI spend digestion rather than chasing strength.

LMND is the most asymmetric but also the most timing-sensitive. The setup depends on operating leverage continuing to outrun growth, which is plausible only if loss severity stays contained and model improvements remain stable through a full claims cycle. In insurance, the consensus often overweights policy growth and underweights catastrophe frequency, so the real question is whether the platform’s unit economics survive a bad weather/claim environment without forcing a reset in underwriting assumptions.

Contrarianly, the market may be underestimating the breadth of winners outside the obvious names: the picks-and-shovels layer around advanced packaging, cloud networking, and power/cooling could capture more durable margin expansion than the headline beneficiaries. The main reversal catalyst is a capex pause, which would show up first in hyperscaler commentary before it hits TSM’s reported numbers by one to two quarters. For LMND, any spike in combined ratio or a miss on the profitability timeline would likely compress the multiple faster than the business would deteriorate, making it a high-beta event-driven name rather than a pure long-term compounder.

Key Decisions for Investors

• Long TSM on any post-earnings consolidation over the next 2-6 weeks; risk/reward favors buying dips because the market is likely to re-rate durable AI capex visibility before it re-prices margin pressure.
• Add AMZN vs. QQQ as a relative-long pair for 3-6 months; AWS acceleration plus ad/fulfillment operating leverage should outperform broad tech if the market rotates toward earnings quality.
• Initiate a small, size-limited long LMND call spread for 3-9 months; upside is large if profitability milestones hold, but keep notional low because underwriting volatility can gap the stock on any claims shock.
• Use a paired hedge: long TSM / short a basket of legacy hardware or semi-capex suppliers with weaker AI exposure over 1-2 quarters; the thesis is that advanced-node scarcity and mix improvement remain concentrated in the highest-quality foundry.
• Set a catalyst alert on hyperscaler capex guidance over the next earnings cycle; any sign of budget normalization is a trigger to trim TSM and AMZN beta exposure quickly.