This Artificial Intelligence (AI) Stock Just Hit an All-Time Low, But Wall Street Says It's Time to Buy

SentinelOne shares hit an all-time low, but the article argues the selloff is overdone and notes 40 covering analysts with 22 buy ratings, 3 overweight ratings, and no sells. The consensus price target is $18.64, about 56% above the current level, with a Street-high target of $26 implying 117% upside. The company reported 2025 revenue growth of 22% to more than $1 billion, an adjusted profit of $68 million, and guided to 20% revenue growth in 2026, supporting the bullish case.

Analysis

The market is conflating two very different risks: AI as a detection layer versus AI as an end-to-end security platform. The former is commoditizing faster than the latter, which means the biggest pressure is likely not on the category leader with a full-stack installed base, but on smaller point-solution vendors whose value proposition is narrower and easier to replicate. In that setup, the near-term winners are incumbents that can bundle agent governance, identity, endpoint, and cloud controls into one workflow, while the long-term threat is margin compression if AI lowers switching costs across the software stack.

The selloff in S looks more like a positioning event than a fundamentals break. If organic growth stays near the low-20s while expense growth remains mid-teens, the market can re-rate the name quickly because the first inflection investors want is not revenue acceleration but proof of durable operating leverage. The key second-order effect is that every increment of AI-agent adoption increases the addressable security surface, which should actually expand wallet share for vendors that can police model behavior, not shrink it.

Consensus may be underestimating how asymmetric this setup is over the next 3-6 months. If the new AI security tools disappoint in real-world deployment, the current de-risking in cybersecurity could unwind violently as investors rotate back to cash-generating incumbents; if they work, the best beneficiaries may still be the established platforms that can integrate them fastest. The bigger risk to the long thesis is not technical obsolescence, but a prolonged reset in enterprise IT spending that delays seat expansion and keeps multiple compression in place.

Relative value still favors S over the larger platform peers on a growth-adjusted basis, but that advantage only matters if execution continues to improve. The cleanest expression is to own the laggard on valuation and fund it against the names least likely to lose strategic relevance, since any broad cyber re-rating will likely lift all boats but the cheapest name should have the highest beta to a sentiment reversal.

Key Decisions for Investors

• Go long S for a 3-6 month mean-reversion trade; use the recent low as a tactical entry, with upside to a re-rating if AI-disruption fears fade. Risk: a real enterprise pilot failure by a major AI-native security entrant could keep multiples depressed.
• Pair trade: long S / short a basket of higher-multiple cyber leaders (CRWD, PANW, ZS) for relative-value compression if the market stops paying for growth at any price. Target 10-15% spread capture over 1-2 quarters; stop if S underperforms on billings or margins.
• Sell puts or structure bullish call spreads on S into any further weakness over the next 30-45 days, expressing a view that the stock has already discounted the worst-case AI threat. Favor defined-risk structures because sentiment can stay broken even if fundamentals remain intact.
• Avoid shorting the group outright until real-world evidence proves AI-native tools can replace layered enterprise security; the better risk/reward is to fade the reaction, not the category. If anything, use weakness in CRWD/PANW/ZS only if their relative multiples re-extend materially without corresponding execution upside.