The Best Artificial Intelligence (AI) Growth Stocks After the Helium Shock Are the Ones That Don't Need the Strait of Hormuz

Datadog reported Q1 2026 revenue of $1.0B, up 32% year over year, with EPS up about 30% to $0.60, free cash flow of $289M, and a raised full-year outlook; shares jumped roughly 30% in the session. CrowdStrike also showed solid AI-driven software demand, with fiscal Q4 ARR of $5.25B, up 24%, revenue of $1.31B, up 23%, and fiscal 2027 revenue guidance of $5.86B-$5.92B. The article argues that AI adoption benefits pure-software observability and cybersecurity platforms that are insulated from physical supply-chain constraints like helium shortages.

Analysis

The market is starting to separate AI beneficiaries into physical-stack winners and software-stack toll collectors. The important second-order effect is that as AI workloads scale, observability and security spend should rise faster than raw compute spend because every incremental model deployment creates continuous monitoring, audit, and incident-response demand. That makes DDOG and CRWD less dependent on the capex cycle than semis, but more exposed to the pace of enterprise AI production rollouts than headline model adoption.

DDOG looks like the cleaner near-term re-rating candidate because the revenue inflection is being driven by a new workload category rather than just share gains in an existing one. The risk is that the current move already prices in a strong AI monetization curve; if GPU monitoring and AI observability fail to expand ARPU over the next 2-3 quarters, the multiple can compress quickly given its premium growth valuation. The better way to own it may be through upside structures rather than outright exposure.

CRWD has a stronger strategic moat but a messier path to multiple expansion because enterprise buyers still carry memory of operational risk and procurement teams will press on pricing after a large run. The underappreciated bullish catalyst is that agentic AI should create more endpoints, identities, and machine-to-machine permissions, which expands the security surface area faster than traditional seat-based security budgets. That said, the market is likely overestimating how quickly this translates into re-accelerating growth; the stock needs proof of renewed net new ARR strength over the next two earnings prints.

The contrarian read is that the article is directionally right but possibly early on the magnitude: AI software demand is real, but the first beneficiaries may be cloud infrastructure and data tools tied directly to inference spend, while security/observability monetization lags by a few quarters. If AI adoption pauses or enterprises continue to test rather than productionize, both names could stall despite favorable secular framing. For now, the trade is about owning the picks-and-shovels layer with lower supply-chain risk, not assuming immediate upside from every AI headline.

Key Decisions for Investors

• Buy DDOG on 2-3 week pullbacks after the post-earnings gap, targeting the next quarterly print; use a 6-12 month horizon and prefer a call spread over common stock to limit multiple compression risk.
• Initiate a CRWD starter long only on evidence of continued net new ARR re-acceleration; if buying now, hedge with a short-dated call overwrite because the stock likely needs another beat-and-raise before the market pays for 20%+ growth again.
• Pair trade: long DDOG / short a basket of semis most exposed to physical bottlenecks (e.g. NVDA/INTC relative under a supply-chain stress scenario) for a 1-3 month window; thesis is that software spend is less sensitive to the geopolitical shock than hardware demand.
• For investors already long both names, trim 20-30% into strength and replace with bull call spreads 1-2 expiries out; risk/reward is better if the goal is participation in AI software adoption without paying peak multiples.
• Set a catalyst watch on the next two enterprise budget cycles: if AI-related observability and security budgets are not clearly separated from generic cloud spend by then, cut exposure, as the rerating thesis depends on budget line-item creation rather than narrative alone.