The Best AI Stocks Wall Street Is Sleeping On in 2026

The article argues that Figma, SentinelOne, and Meta Platforms may be undervalued AI-related opportunities, highlighting Figma's 136% net dollar retention, 40% Q4 revenue growth to $304 million, and $38 million in positive free cash flow. SentinelOne posted 22% ARR growth and 20% Q4 revenue growth to $271 million, while Meta grew Q1 revenue 33% to $56.3 billion and generated $12.4 billion in free cash flow despite raising 2026 capex to $125 billion-$145 billion. Overall, the piece is bullish on AI stocks with depressed valuations rather than signaling a near-term catalyst.

Analysis

The common thread here is not “AI winners” so much as a repricing of software duration risk. The market is starting to distinguish between companies where AI is a feature that can be commoditized and companies where AI is embedded in workflow, data, or distribution moats. That favors Meta most clearly: its ad engine can monetize AI immediately, while its data advantage makes third-party model displacement much less plausible than in horizontal software.

Figma and SentinelOne are being treated as if AI is primarily a competitive threat, but the second-order effect is more nuanced: AI lowers acquisition friction for the category leader and raises the cost of switching for serious enterprise users. If the core user base values precision, governance, and integration, cheaper AI clones may expand the low end of the market without compressing the high end as much as bears expect. The key risk is not product obsolescence; it is a slower-than-expected reacceleration in net-new customer adds, which would keep multiples pinned for 2-4 quarters.

The contrarian opportunity is that valuation compression has already priced in a lot of “AI disruption,” especially for FIG and S, while balance-sheet and cash-flow trajectories are improving. Meta is the cleaner quality-growth expression, but it is also the one where capex scrutiny can create periodic air pockets; those are likely buying opportunities if ad growth stays north of the low-20s. In other words, the market is overpaying for obvious AI infrastructure beneficiaries and underpricing incumbents that can self-fund AI monetization.

For the broader complex, this setup is a relative-value rotation, not a sector-wide long. If investor sentiment rotates from narrative to fundamentals, the next leg should favor profitable compounders with durable demand and punish names whose AI story is still mostly optionality.

Key Decisions for Investors

• Long META on pullbacks over the next 1-3 weeks; target a 15-20% upside move over 6-9 months if revenue growth holds above ~25% and capex fears ease. Risk/reward is attractive because downside is cushioned by cash flow, while multiple expansion can resume quickly.
• Initiate a FIG/S basket long against a broad software short or against ADBE if liquidity allows; hold 3-6 months. The thesis is that AI commoditization fear is over-discounted in the two names with improving retention and cash generation, while ADBE remains the cleaner implied-disruption short on distribution and legacy workflow exposure.
• Buy 3-6 month call spreads on META into any post-capex selloff. Use defined risk because the catalyst is not immediate product news, but a re-rating once investors see capex convert into engagement and ad yield rather than margin collapse.
• Avoid chasing NVDA here unless the entry is tied to a specific earnings re-acceleration setup. The market is already paying for perfection, and the article’s signal is valuation dispersion within AI, not another leg of indiscriminate AI beta.
• For higher-beta expression, pair long S with a short in a slower-moving legacy cyber name over 1-2 quarters. The setup is that AI-native security vendors can win share even in a cautious spend environment, but position sizing should be modest because the path to sustained profitability remains the key risk.