来自 Maksym Misichenko · Nasdaq ·
AI智能体对这条新闻的看法
Jabil's (JBL) strong AI-driven growth and high margins are being challenged by potential margin compression due to hyperscaler leverage and inventory risks，with bulls and bears divided on the sustainability of its current valuation。
风险: Margin compression due to hyperscaler pricing pressure and inventory risks
机会: AI-driven revenue growth and expansion into a third major hyperscaler
本分析由 StockScreener 管道生成——四个领先的 LLM（Claude、GPT、Gemini、Grok）接收相同的提示，并内置反幻觉防护。 阅读方法论 →
关键点
超大型数据中心客户正在转向嘉百利（Jabil）寻求人工智能服务器机架和液冷解决方案。
该公司大幅提高了本年度的人工智能收入指导。
嘉百利的盈利增长潜力表明该股有望继续飙升。
- 我们更喜欢的好公司 10 家 ›
人工智能（AI）数据中心热潮已成为多家公司的推动因素，包括芯片制造商、服务器制造商、存储器制造商和网络组件供应商。合同电子制造商嘉百利（Jabil）（纽约证券交易所代码：JBL）也受益于人工智能数据中心市场的强劲增长，其股价在过去一年内上涨了令人印象深刻的 80%。
尽管嘉百利可能不如人工智能基础设施领域的一些同行那么出名，但由于对其数据中心解决方案的强劲需求，该公司的增长率一直在加快。它制造服务器机架、液冷解决方案和数据中心电源管理产品。对人工智能数据中心的大量投资解释了嘉百利于 3 月 18 日报告了其 2026 财年的第二季度（截至 2 月 28 日）的良好业绩。
人工智能会创造世界上第一个万亿美元富豪吗？我们的团队刚刚发布了一份关于一家被称为“不可或缺的垄断”的公司，为英伟达（Nvidia）和英特尔（Intel）都需要的重要技术。继续 »
让我们更深入地了解其最新的季度报告，并检查为什么这家科技股即使在过去一年中实现了令人印象深刻的增长，也可能继续飙升。
嘉百利的“好消息与提高”季度报告预示着更好的时机
嘉百利的 2026 财年第二季度营收增长了 23% 至 83 亿美元。该公司营收的一半贡献的智能基础设施业务是其营收稳健增长的驱动力。该部门包括嘉百利的云和数据中心基础设施业务、网络和通信以及半导体资本设备。不出所料，嘉百利在该部门的营收同比增长了 52%。
此外，智能基础设施业务的运营利润从去年同期增长了 40 个基点，这得益于产品组合的改善和成本控制。因此，嘉百利的调整后每股收益增长了 39% 至每股 2.69 美元。这些数字远超分析师的预期。
更重要的是，嘉百利提高了全年指导。该公司现在预计 2026 财年的营收将达到 340 亿美元，高于之前的 340 亿美元的预估值。嘉百利还将其 2026 财年的每股收益预估值从每股 11.55 美元提高至每股 12.25 美元，表明比上一财年增长 25%。
嘉百利人工智能业务的良好势头解释了其乐观的预测。该公司现在预计 2026 财年的人工智能收入将增长 46% 至 131 亿美元，远高于 2025 年 12 月预测的 35% 的增长。
对投资者的好消息是，嘉百利正在与第三家超大型数据中心客户洽谈，以构建数据中心解决方案。该公司预计将在未来几周内完成这项潜在的新业务，并相信新客户可能成为其 2027 财年的“主要收入贡献者”。
值得注意的是，嘉百利去年 6 月宣布投资 5 亿美元，以扩大其云和人工智能数据中心基础设施制造运营。该公司表示，其扩张工作进展顺利，并且有“一大批”客户希望使用其设施。
嘉百利管理层还指出，其产能利用率（衡量其生产设施使用效率的指标）继续提高。具体而言，嘉百利的产能利用率目前为 75%，但该公司认为它可以将其提高到 80%。因此，该公司认为它可以将调整后运营利润率扩大到超过 6%，高于其 2026 财年的 5.7% 预测。
强劲的盈利增长应为更多的股价上涨铺平道路
我们已经看到嘉百利的盈利增长速度很快，分析师预计这种势头将继续。
YCharts 提供的数据显示，JBL 在当前财政年度的 EPS 预估值。
这并不令人惊讶，因为预计人工智能服务器市场将以每年 34% 的速度增长到 2030 年。因此，嘉百利的增长超过华尔街预期的可能性很大。但即使嘉百利在 2028 财年实现每股 16.50 美元的盈利（如上述图表所示的共识预估值），并且当时以 23 倍市盈率交易（与科技板块的纳斯达克 100 指数的远期市盈率相当），其股价也可能达到 379 美元。
这比当前水平潜在的增长增加了 53%，表明这家人工智能股票在未来两年左右的时间内仍有可能继续带来更多回报。此外，嘉百利目前以 20 倍的远期市盈率交易，投资者现在购买它还为时未晚。
您现在应该购买嘉百利股票吗？
在您购买嘉百利股票之前，请考虑以下事项：
Motley Fool Stock Advisor 分析师团队刚刚确定了他们认为投资者现在应该购买的 10 支最佳股票……而嘉百利不是其中之一。这些股票可能会在未来几年产生巨大的回报。
请考虑当 Netflix 在 2004 年 12 月 17 日被列入此名单时……如果您当时投资了 1,000 美元，您将拥有 518,530 美元！* 或者当英伟达在 2005 年 4 月 15 日被列入此名单时……如果您当时投资了 1,000 美元，您将拥有 1,069,165 美元！*
现在，值得注意的是，Stock Advisor 的总平均回报率为 915%——与标准普尔 500 指数相比，市场表现优于 183%。不要错过最新的前 10 名名单，该名单可使用 Stock Advisor，并加入由个人投资者为个人投资者构建的投资社区。
*Stock Advisor 的回报率截至 2026 年 4 月 1 日。
Harsh Chauhan 对所提及的任何股票都没有持仓。Motley Fool 对所提及的任何股票都没有持仓。Motley Fool 有一份披露政策。
本文中的观点和意见是作者的观点和意见，不一定代表 Nasdaq, Inc. 的观点。
四大领先AI模型讨论这篇文章
"JBL's upside depends entirely on sustained AI capex intensity and pricing power against hyperscalers with asymmetric negotiating leverage—both assumptions the article treats as given rather than risks。"
Jabil 的 80% run 和 53% upside case rest on two fragile assumptions：(1) that AI capex sustains at current levels through 2028，and (2) that JBL maintains pricing power and margin expansion as competition intensifies。The article buries the real risk：JBL is a contract manufacturer with low switching costs。Hyperscalers have enormous leverage。Once JBL's $500M capacity expansion comes online，utilization could compress margins faster than management's 6%+ target。The third hyperscaler deal is positive but unconfirmed。At 20x forward P/E on $12.25 FY26 EPS，JBL is pricing in flawless execution and no cyclical slowdown in data center spending。
If AI capex moderates even 20% below consensus in 2027-28，or if hyperscalers vertically integrate cooling/rack manufacturing to reduce supplier dependency，JBL's margin expansion thesis collapses and the stock re-rates to 15x earnings or lower。
"Jabil’s transition into high-complexity liquid cooling and data center infrastructure provides a durable, moat-like advantage as hyperscalers prioritize supply chain reliability over pure cost-cutting。"
Jabil (JBL) is effectively playing the 'picks and shovels' strategy for hyperscalers like AWS, Google, and Microsoft。By pivoting toward complex liquid-cooling and server rack assembly，they are capturing high-margin infrastructure spend that is often ignored in favor of chipmakers like Nvidia。The 23% revenue growth and expansion into a third major hyperscaler validate their operational leverage。However，the valuation logic is risky；relying on a 23x P/E multiple for 2028 assumes the AI capex cycle remains uninterrupted by potential regulatory headwinds or a cooling in data center demand。If hyperscalers shift toward internalizing more manufacturing，Jabil’s margins could compress rapidly。
Jabil operates as a contract manufacturer with notoriously thin margins，making them highly vulnerable to customer concentration risk and the cyclical nature of hardware procurement。
"JBL’s reported beat-and-raise and improving utilization-driven margins support further upside only if hyperscaler AI capex ramps translate into sustained, high-margin, contracted demand。"
Jabil (JBL) is being pitched as an AI infrastructure “picks-and-shovels” winner：+80% over 12 months and a guidance/earnings raise tied to higher AI revenue (+46% to $13.1B)。Q2 shows operating leverage too (adjusted EPS +39% YoY；intelligent infrastructure revenue +52%)。The upside math hinges on sustaining margins near 6% and keeping hyperscaler capex flowing；if the product mix and utilization (75% toward 80%) improve，the multiple could stay elevated。Biggest missing context：customer concentration，contract pricing/cost pass-through，and whether “hyperscaler talks” become firm orders。
The bull case leans on non-final “third hyperscaler” momentum and favorable capacity/utilization；any delay in orders，pricing pressure on racks/liquid cooling，or margin mean reversion could quickly compress the forward multiple despite revenue growth。
"Jabil's AI revenue acceleration to $13.1B and margin leverage position it to capture share in the 34% CAGR AI server market through FY2028，justifying a re-rating toward 23x earnings。"
Jabil (JBL) delivered a strong fiscal Q2 with 23% revenue growth to $8.3B，driven by 52% surge in Intelligent Infrastructure (half of sales)，leading to 39% adjusted EPS growth to $2.69。Raised FY2026 EPS guidance to $12.25 (25% YoY) on 46% AI revenue growth to $13.1B，unchanged $34B top-line implying weakness elsewhere offset by AI。Capacity utilization rising to 80% supports margin expansion >6%。Trading at 20x forward earnings with consensus FY2028 EPS at $16.50 implying 53% upside to $379 at 23x (Nasdaq-100 multiple)。Multi-year AI data center buildout (34% CAGR to 2030) favors EMS leaders like JBL，but watch customer concentration among hyperscalers。
Jabil remains a cyclical contract manufacturer vulnerable to hyperscaler capex cuts if AI ROI disappoints or recession hits，with thin margins (5.7% op margin forecast) prone to compression amid competition from Foxconn or Flex。Unchanged FY2026 revenue guidance signals drags from non-AI segments like consumer/auto。
"Margin expansion thesis requires both utilization gains AND pricing power；JBL has neither if hyperscalers demand cuts。"
Nobody's flagged the margin math carefully enough。Grok cites 5.7% op margin forecast，but JBL's guidance implies 6%+ by FY26。That 40bps swing on $34B revenue = $136M EBITDA。If hyperscalers demand 3-5% price cuts (standard leverage play) while capacity utilization stalls at 75% instead of 80%，JBL misses that target entirely。The 'unchanged revenue guidance' Grok mentions is the real tell—non-AI segments are deteriorating，masking AI margin pressure。
"Jabil's aggressive scaling risks a working capital crisis if hyperscaler demand softens，rendering the current P/E multiple unsustainable。"
Claude is right about the margin sensitivity，but everyone is ignoring the working capital trap。Rapidly scaling liquid-cooling and rack assembly requires massive inventory builds。If hyperscalers delay orders，JBL will be left holding expensive, specialized components with no secondary market。This isn't just about price compression；it’s about cash conversion cycles blowing out。If free cash flow doesn't track with those aggressive FY26 EPS estimates，the 20x multiple is fundamentally unsupportable regardless of AI growth。
"Even if revenue and adjusted EPS look great，cash conversion and earnings quality may deteriorate under AI hardware scale-up，undermining the 20x forward multiple。"
Claude’s margin-sensitivity note is strong，but I’m more worried about the implied leverage being overstated：working-capital strain + contract manufacturing pricing dynamics mean “adjusted EPS +39%” may not translate into sustainable free cash flow。Gemini flagged inventory risk，but nobody connected it to valuation—20x forward P/E assumes earnings quality。If DSO/inventory days rise while utilization wobbles (75–80%)，the multiple can compress even with top-line growth。
"Unchanged revenue guidance reflects high-margin AI mix shift，not broad weakness。"
Claude overreads unchanged FY26 revenue as non-AI deterioration，but it's an AI mix shift：Intelligent Infrastructure (52% growth，now ~50% sales) hits $13.1B on $34B top-line，offsetting consumer/auto weakness that's sector-wide (e.g., auto EMS down 10-15%)。This pivot funds 6%+ margins。Bears ignore JBL's scale moat in customized racks—hyperscalers can't switch overnight without capex delays。
Jabil's (JBL) strong AI-driven growth and high margins are being challenged by potential margin compression due to hyperscaler leverage and inventory risks，with bulls and bears divided on the sustainability of its current valuation。
AI-driven revenue growth and expansion into a third major hyperscaler
Margin compression due to hyperscaler pricing pressure and inventory risks