Ontology of Power
In this edition of the Smart Investor newsletter, we spotlight the AI operating layer connecting data to action. But first, let’s review the latest news and developments.
1
Portfolio News and Updates
❖❖ NVIDIA (NVDA) used its GTC Taipei / Computex 2026 keynote to make its boldest move yet beyond the data center space. CEO Jensen Huang’s message was that NVIDIA intends to supply the entire computing stack, from the personal computer to the hardware and software behind the largest AI data centers.
The headline was NVIDIA’s entry into the PC market. Its new RTX Spark superchip pairs an Arm-based CPU with a Blackwell GPU on TSMC’s (TSM) 3nm process, delivering a full petaflop of AI performance and up to 128GB of memory in a laptop. Built in partnership with Microsoft (MSFT) for a new generation of Windows laptops and compact desktops arriving this fall, RTX Spark is designed to run AI agents locally, turning the PC from a tool you operate into one that can work on your behalf.
With this superchip, NVIDIA is pushing directly into the personal computer processor market for the first time, opening an enormous new addressable opportunity beyond its data center base. The catch is that early systems target the premium end, pricing is still unannounced, and the ongoing memory shortage could make high-spec models expensive – so this is a long-term strategic wedge, not an immediate revenue driver.
The actual near-term engine still sits in the data center. NVIDIA confirmed its next-generation Vera Rubin platform – the successor to Blackwell – is ramping into full production, built for the agentic AI workloads where a single prompt can trigger thousands of reasoning steps.
Crucially, NVIDIA also launched Vera, its first CPU designed specifically for AI agents, claiming 1.8x faster performance than traditional x86 chips. OpenAI, Anthropic and SpaceX are among the first named customers exploring the Vera CPU. The AI leader previously signaled that CPUs are one of its key targets, and now it’s showing how it plans to harness that $200+ billion market opportunity.
Beside the enormous additional revenue runway, the CPU entry completes NVDA’s strategy to become the full-stack AI king. By adding its own CPU to its GPUs, networking and software, NVIDIA extends control over the entire AI infrastructure layer just as agentic AI drives a new wave of compute demand.
Around these pillars, NVIDIA reinforced its moat with a software push for autonomous “AI coworker” agents, a new DSX AI factory playbook, and a sweeping physical-AI expansion: the Cosmos 3 robotics model, an open humanoid-robot reference design, and a broadened DRIVE Hyperion robotaxi platform with partners including Foxconn and Uber. These address enormous long-term markets but sit further out on the horizon.
The bottom line: the market’s attention landed on the PC move, but the heavier weight is Vera Rubin entering full production and NVIDIA tightening its grip on every layer of the AI economy.
1
❖❖ IBM’s (IBM) stock soared last week on the Trump administration’s $1 billion quantum award, and continued climbing this week following the company’s massive $15+ billion combined commitment to quantum computing and AI-driven open-source security.
The bulk of that spend targets quantum. Over the past two weeks, IBM secured the $1 billion federal grant to build Anderon – America’s first pure-play quantum foundry – matching it with $1 billion of its own funding. The tech giant also committed more than $10 billion over five years to reach the field’s holy grail: the first large-scale fault-tolerant quantum computer by 2029. IBM already touts over 90 deployed systems, claiming more quantum computers than all other players combined.
The other $5 billion goes to Project Lightwell, an IBM and Red Hat initiative deploying 20,000 engineers and AI to secure the open-source software that underpins modern enterprise and AI systems. Early adopters span Wall Street’s biggest names, including JPMorgan (JPM), Goldman Sachs, and Visa.
Barclays’s initiation with a Buy rating and a $350 price target confirmed the market’s optimism, giving the rally another push. The firm highlighted IBM’s sticky software business, regulated enterprise focus, and insulation from AI disruption – a setup for compounding earnings growth even if quantum ambitions are pushed out further in time.
Meanwhile, Wedbush pinpoints its optimism on IBM’s combined AI and quantum positioning, calling it a “$1 trillion value creation engine” that remains underappreciated. Analysts believe quantum advantage – the point at which quantum systems meaningfully outperform classical computers on practical tasks – could arrive sooner than many expect. IBM has long positioned itself for this shift, including through quantum-related security initiatives like IBM Quantum Safe. Together with the Lightwell project, this creates a compelling full-stack strategy: IBM aims to serve as the trusted provider for both building advanced AI/quantum systems and securing them.
1
❖❖ Synopsys (SNPS) delivered a strong fiscal Q2 2026 beat-and-raise, but the bigger story was the market’s reassessment of its role in the AI infrastructure stack. Revenue surged 42% year-over-year to $2.28 billion, while non-GAAP EPS came in at $3.35, both ahead of expectations. Management also raised full-year guidance, now expecting revenue of $9.63-9.71 billion – up about 37% year-over-year at the midpoint – and non-GAAP EPS of $14.72-14.80, up roughly 14%, while lifting its non-GAAP operating margin target to about 41%, a nearly 370-basis-point expansion from FY2025.
The quarter reinforced the argument that AI is driving demand across the entire SNPS portfolio. Design Automation revenue climbed to $1.82 billion, supported by advanced-node design activity, 3DIC adoption, hardware-assisted verification demand, and growing interest in AI-powered engineering tools. The company highlighted more than 30 advanced-node technical wins, continued chiplet and advanced packaging momentum, and strong early traction for Multiphysics Fusion and agentic EDA solutions. Management disclosed that 20 customers are evaluating more than 25 specialized AI agents across verification, implementation, and analog design workflows.
Just as important, Synopsys is becoming more deeply embedded in the AI ecosystem. Shortly after earnings, the company expanded its Samsung Foundry collaboration around AI-powered 2nm and multi-die design flows, while NVIDIA (NVDA) CEO Jensen Huang identified Synopsys as one of the first partners building autonomous AI engineers using NVIDIA’s NemoClaw platform. That endorsement strengthened the case that SNPS is moving beyond traditional EDA into a broader AI-driven engineering platform.
Ansys is central to that shift. Beyond its near-term revenue contribution, the acquisition expands Synopsys from chip design into system-level simulation, multiphysics analysis, packaging, thermal modeling, and full silicon-to-systems engineering – exactly where AI infrastructure complexity is rising fastest. Management said integration is progressing ahead of schedule, expects roughly half of targeted cost synergies to be realized by fiscal year-end, and ended Q2 with an $11 billion backlog.
Meanwhile, activist investor Elliott Management reached a cooperation agreement with Synopsys, resulting in Managing Partner Jesse Cohn joining the board. This adds another experienced voice to Synopsys’ leadership as the company navigates the Ansys integration and expands its AI-related opportunities. His appointment reinforces sharper execution, faster margin capture, and a clearer translation of AI demand into shareholder value.
While concerns around Ansys integration and weakness in the Design IP segment initially pressured shares following earnings, sentiment has since improved amid analyst price-target increases, a broader software rally, and growing confidence that AI-driven demand, Ansys synergies, and new monetization opportunities can support sustained growth and margin expansion.
1
❖❖ Credo Technology (CRDO) stock dropped following its fiscal Q4 2026 report despite delivering another blowout quarter, as investors appeared to lock in gains after a remarkable pre-earnings rally. The stock had surged more than 160% from its March lows heading into the print, creating an exceptionally high bar even for a company that continues to outperform expectations.
The results themselves were difficult to fault. Q4 revenue jumped 157% year-over-year to a record $437 million, exceeding expectations and surpassing the company’s entire fiscal 2025 revenue. Non-GAAP EPS came in at $1.16, up 231% year-over-year and well above consensus, while non-GAAP gross margin reached 68.3% and non-GAAP net margin exceeded 50%. For full fiscal year 2026, revenue more than tripled to $1.34 billion, non-GAAP net income increased more than fivefold to $662 million, and EPS surged 392% to $3.46.
CRDO guided above expectations for FQ1, forecasting revenue of $465-475 million. More importantly, the company outlined a dramatically stronger fiscal 2027 outlook than investors had been discussing just a few months ago. Credo now expects more than 80% revenue growth this year, supported by a major optical networking ramp that management believes will generate more than $600 million in revenue. Optical DSPs, silicon-photonic PICs, and ZeroFlap Optics are each expected to contribute more than $100 million, with the acceleration weighted toward the second half of the year. The recently completed Dust Photonics acquisition further strengthens that roadmap.
The results also reinforced the core investment thesis behind CRDO’s recent re-addition to the Smart Investor Portfolio. As AI clusters grow larger and more complex, connectivity is becoming a critical bottleneck alongside compute. Credo’s Active Electrical Cables remain a major growth driver, while optics, retimers, and emerging memory-connectivity solutions expand the company’s addressable market and deepen its role inside AI infrastructure.
Although the stock initially dropped about 11-12% in after-hours trading following the earnings release, buyers quickly stepped in and recovered much of the decline. By the next day, the stock returned to gains, indicating that the reaction was likely driven more by profit-taking after the stock’s huge pre-earnings rally than by concerns about the company’s fundamentals. Given CRDO’s volatility and strong recent performance, that distinction is important.
1
❖❖ Palo Alto Networks (PANW) delivered another beat-and-raise quarter, as AI-driven security demand continues to accelerate across its platform portfolio. Fiscal Q3 2026 revenue rose 31% year-over-year to a record $3.0 billion, while adjusted EPS came in at $0.85, ahead of expectations. PANW has now beaten revenue estimates for five straight quarters, while its string of adjusted EPS beats stretches back to at least 2021. The headline, however, isn’t another beat but acceleration – especially in the areas most important to both short- and long-term growth.
AI is now a major growth catalyst for cybersecurity leaders – fundamentally changing enterprise security requirements and creating new demand – and Palo Alto’s numbers show it. In contrast to most cybersecurity companies, PANW has several AI platforms, with each of them boosted by AI simultaneously. AI data centers are boosting firewall demand, AI adoption is driving Prisma AIRS, AI-generated telemetry is helping XSIAM and Observability, and AI agents create a new identity-security opportunity for CyberArk.
As a result, Next-Generation Security (NGS) ARR – Palo Alto’s key metric tracking recurring revenue from its next-gen security portfolio – soared 60% year-over-year, nearly double the previous quarter’s growth, reaching $8.13 billion. Importantly, NGS ARR excluding CyberArk and Chronosphere still grew 28% year-over-year. One of the most notable developments was Prisma AIRS, the company’s dedicated AI security platform protecting the entire AI lifecycle, from model development through runtime. The platform is less than a year old, yet its customer count jumped threefold compared to Q2, and management expects AIRS to exceed $100 million in ARR within the next couple of quarters. Management said it is the fastest-scaling product in company history.
Alongside new and exciting products, the longstanding pillar of PANW’s business, Network Security, surprised to the upside, also supported by AI-related demand. Next-gen firewall bookings surged nearly 40% year-over-year, producing the segment’s strongest third quarter in years. AI data centers, rising machine-to-machine traffic, and agentic AI applications are turning network inspection back into a multi-year growth engine.
The platformization strategy also continued gaining traction. Palo Alto added 110 net-new platformized customers during the quarter, bringing the total to 2,280. Around 65% of NGS ARR comes from platformized customers, who generate roughly 120% net retention rates and higher spending levels.
CyberArk and Chronosphere integrations are tracking ahead of schedule, with both acquired assets performing better than expected, directly addressing investors’ integration worries. CyberArk contributed more than $1.3 billion of NGS ARR, while Chronosphere surpassed $300 million in ARR. Both platforms also materially boosted the company’s order backlog, with total RPOs jumping 36% year-over-year to $18.4 billion. Current RPO was $8.3 billion, up 34%.
PANW guided Q4 revenue to $3.345-3.355 billion, implying roughly 32% year-over-year growth at the midpoint, with adjusted EPS of $0.96-0.98 – both well above Street expectations. Moreover, NGS ARR is expected to come in at $8.9-8.95 billion, reflecting 59-60% growth, while the projected RPO of $20.9-21 billion implies 32-33% growth.
The company raised its FY 2026 guidance across all metrics, driven by accelerating organic bookings and strong integration of recent acquisitions. Management now expects revenue of $11.415-11.425 billion, up 24%; adjusted EPS of $3.77-3.79; Next-Generation Security ARR growth of about 60%; RPO growth of about 33%; non-GAAP operating margin of 28.9-29.2%; and adjusted free cash flow margin of 37.5%. Overall, Palo Alto’s results and raised guidance reinforced its leadership position and wide moat, showing that the company is effectively harnessing secular AI and cybersecurity tailwinds across its expanding platform portfolio.
1
Portfolio Earnings and Dividend Calendar
❖ The Q1 2026 earnings season has ended, but several Smart Investor Portfolio holdings with fiscal years different from the calendar ones are yet to reveal their latest quarterly results. Broadcom (AVGO) and CrowdStrike Holdings (CRWD) are scheduled to release their reports today after the close. Their results are highly impactful, as AVGO is the semiconductor cycle bellwether, while CRWD is the cybersecurity read-through.
❖ The ex-dividend date for NVIDIA (NVDA) is June 4, while for Alphabet (GOOGL) it is June 8.
w
1
New Buy: Palantir Technologies (PLTR)
Palantir Technologies, Inc. sits at the center of one of the most important shifts in enterprise technology: turning artificial intelligence from a promising tool into an operational system. Its platforms do not simply store data or display dashboards – they connect fragmented information, business logic, security permissions, and AI models into a working layer that helps organizations make decisions and act on them. That role is becoming more valuable as governments and enterprises move beyond AI pilots and seek software that can run inside real-world operations. As AI adoption accelerates, Palantir is positioning itself as a critical infrastructure provider, enabling organizations to integrate data, automate workflows, and deploy AI-driven decision-making at scale.
1
Domination Sequence
Palantir Technologies was founded in 2003 by Peter Thiel, Alex Karp, Stephen Cohen, Joe Lonsdale, and Nathan Gettings with the idea that organizations should be able to extract value from vast amounts of data without sacrificing security or control. The company’s early years were shaped by work with U.S. intelligence and defense agencies, where complex missions, fragmented information, adaptive adversaries, and strict access requirements created problems that conventional software struggled to solve. Over time, PLTR established itself as a trusted technology partner across national security, law enforcement, and government organizations, building a reputation for tackling high-stakes operational challenges.
The more transformative chapter began over the past six years. Following its 2020 direct listing, Palantir accelerated its push into the commercial market, applying capabilities developed in government environments to large enterprises facing increasingly complex operational and data-management challenges. The company expanded across industries including healthcare, manufacturing, energy, transportation, and financial services, broadening its addressable market while reducing dependence on government contracts.
A key turning point came with the rise of generative AI. While much of the industry focused on developing increasingly powerful models, Palantir positioned itself around a different challenge: helping organizations deploy AI inside real-world operations. The launch of its Artificial Intelligence Platform (AIP) in 2023 became a catalyst for adoption, allowing customers to operationalize large language models securely in real-world environments. The company’s AIP Bootcamp strategy further accelerated customer acquisition by shortening deployment cycles and demonstrating practical use cases in days instead of long consulting-style rollouts.
At the same time, Palantir deepened relationships with major technology providers, including strategic collaborations with Microsoft Azure, Oracle Cloud Infrastructure, AWS, NVIDIA, and others. These partnerships expanded distribution channels, strengthened cloud interoperability, and positioned the company to benefit from the rapid growth of enterprise AI spending. PLTR also increased its presence in some high-profile defense modernization initiatives across the U.S. and allied nations, aligning itself with long-term government priorities surrounding AI, battlefield software, and data-driven decision-making.
This evolution has turned Palantir from a specialized government-focused software company into one of the most visible platforms for operational AI and decision intelligence, sitting at the intersection of enterprise operations, national security, and applied artificial intelligence. As organizations increasingly seek to deepen and expand AI integration and execution, PLTR enters its next phase with a larger market opportunity and a stronger strategic position than at any point in its history.
1
The Philosophy of Action
Palantir operates in one of the most important layers of the AI stack – the point where data, software, human decision-makers, and increasingly AI agents converge to run real-world operations. While much of the AI industry focuses on building larger models or deploying more computing power, PLTR addresses a different challenge: turning AI into something organizations can actually use inside mission-critical environments.
At the center of the business is what management increasingly describes as an Enterprise Operating System, built across several interconnected layers. Foundry is the data-operations platform, built to integrate fragmented information and turn it into a usable model of the enterprise. Gotham is the government- and intelligence-focused counterpart. AIP is the AI layer, connecting LLMs and agents to that operational model so they can support real workflows instead of sitting outside them. Apollo is the delivery layer, managing deployment and updates across cloud, on-premise, edge, and highly secure environments. Binding these pieces together is the Ontology – Palantir’s digital map of how an organization functions, linking data, business rules, assets, decisions, permissions, and actions into a single operational system. This architecture allows both humans and AI systems to work from the same source of truth and, importantly, act on it.
The distinction between AI models and the operational systems that deploy them is increasingly important. Many organizations have accumulated vast amounts of data and gained access to powerful AI models, yet struggle to translate either into operational outcomes. Palantir’s value proposition is closing that gap. The platform does not simply analyze information; it helps organizations make decisions, coordinate workflows, and execute actions across complex systems while maintaining governance, auditability, and security. As enterprises move toward agentic AI, the need for this orchestration layer should become even more important. In many ways, PLTR benefits from AI-model commoditization because cheaper and more capable models increase demand for platforms that can securely deploy, govern, and integrate them.
Commercial adoption has become the company’s fastest-growing opportunity. Demand is being driven by organizations seeking to modernize operations, automate workflows, accelerate decision-making, and integrate AI into everyday business processes. AIP Bootcamps have become a key commercial tactic, compressing long enterprise-sales and deployment cycles into hands-on sessions where customers build practical use cases before committing to broader rollouts.
Strategic partnerships are expanding Palantir’s reach well beyond its own sales force. Collaborations with SAP and Accenture position AIP inside large-scale ERP modernization projects, while relationships with Microsoft, AWS, Oracle, NVIDIA, and Dell increasingly place Palantir at the intersection of enterprise software, cloud infrastructure, and AI deployment. Rather than competing directly with these companies, PLTR is increasingly becoming the software layer that connects them. The Dell partnership also gives Palantir a packaged path into regulated and sovereign AI deployments, combining Dell infrastructure and NVIDIA hardware with PLTR’s software layer for customers that need on-premise AI, tighter auditability, and stronger control over sensitive data.
Government remains equally important. Palantir’s software is deeply embedded in defense modernization initiatives across the U.S. and allied nations, including battlefield decision systems, military interoperability programs, logistics, intelligence operations, and AI-enabled command-and-control platforms. The Maven Smart System, growing adoption across Department of Defense programs, and expanding cooperation with allied governments illustrate how the company’s role has evolved from analytics provider to operational infrastructure. Its involvement in Ukraine further highlights this shift, with PLTR’s software supporting intelligence analysis, drone-interception initiatives, and other AI-enabled defense capabilities.
The opportunity from here extends well beyond traditional software spending. Sovereign AI initiatives, defense modernization, industrial automation, AI-driven enterprise transformation, and autonomous systems all require secure platforms capable of connecting data, models, workflows, and decision-makers. Palantir’s expanding ecosystem – spanning governments, manufacturers, aerospace companies, defense tech, cloud providers, infrastructure vendors, and enterprise software leaders – increasingly positions the company as the operational layer through which those systems interact. The challenge will be scaling fast enough to meet demand, particularly as adoption continues to accelerate. For now, that appears to be a favorable problem to have.
1
Mathematics of Intelligence
Palantir’s financial profile increasingly reflects a business operating at a level few companies ever reach, especially those of PLTR’s scale: rapid growth, expanding profitability, substantial cash generation, and customer stickiness – all happening simultaneously. Rapid growth alone is not unusual in AI. What stands out is that PLTR is delivering accelerating revenue, expanding margins, strong free cash flow, and growing customer spending simultaneously.
Q1 2026 was the clearest example yet. Revenue surged 85% year-over-year to $1.63 billion, the fastest growth rate Palantir has reported since becoming a public company. The company also extended a long-running pattern of execution, beating analyst revenue estimates for at least eight consecutive quarters and adjusted EPS expectations for eleven straight quarters dating back to Q3 2023. The latest quarter’s adjusted EPS soared 154% year-over-year to $0.33, another notch in a long string of high double- and low triple-digit increases.
Growth was broad-based, but the U.S. revenue remained the primary engine, rising 104% year-over-year to $1.28 billion and now representing nearly 80% of the total. That included 133% growth in U.S. commercial revenue and 84% growth in U.S. government revenue.
The internal mechanics of that growth may be even more important than the headline numbers. Customer count increased 31% year-over-year to 1,007, while U.S. commercial customers grew 42% to 615. Growth is being driven by both customer acquisition and strong expansion within existing accounts. Thus, revenue from the top 20 customers increased 55% year-over-year to $108 million per customer. Meanwhile, net dollar retention reached 150%, one of the highest levels in the broader enterprise software sector. And all this is happening with just 70 salespeople on its staff.
Total remaining deal value climbed 98% year-over-year to $11.8 billion, while RPOs reached $4.5 billion, up 134%. PLTR signaled that surging demand, particularly in the U.S., continues to outpace recognized revenue. This doesn’t come as a surprise, as operational results provide tangible evidence that customers are generating measurable returns from deployment. Thus, GE Aerospace reported a 26% improvement in engine-production performance using AIP, while the DoD’s Maven Smart System doubled usage over four months and increased fourfold over the past year.
Profitability remains unusually strong for a company growing at this pace. GAAP operating income reached $754 million, representing a 46% margin, while GAAP net income climbed to $871 million, or 53% of revenue. Adjusted gross margin reached 88% and adjusted operating margin reached 60%. Cash generation was equally impressive, with $899 million in operating cash flow and $925 million in adjusted free cash flow during the quarter. Meanwhile, the company carries minimal debt and has no outstanding borrowings under its credit facility. Put simply, Palantir is not sacrificing profitability or clean balance-sheet to fund growth; it is generating substantial profits without leverage – while still expanding aggressively.
Despite notching meteoric growth over multiple quarters, PLTR doesn’t plan to rest on its laurels, continuing to invest heavily in business expansion and tech efforts. Adjusted operating expenses rose 32% year-over-year to $649 million as the company expanded technical hiring and accelerated development of AIP and related capabilities. Stock-based compensation also remains meaningful at $202 million during the quarter. Even so, revenue growth far outpaces expense growth, driving increasing operating leverage.
The outlook remains equally strong. Management raised full-year 2026 revenue guidance to $7.65-7.66 billion – well above Wall Street expectations – implying approximately 71% year-over-year growth. Adjusted income from operations is expected to reach $4.44-4.45 billion, while adjusted free cash flow is projected at $4.2-4.4 billion. U.S. commercial revenue alone is expected to exceed $3.22 billion, implying at least 120% growth. For Q2, Palantir guided revenue to roughly $1.80 billion at the midpoint (up about 80% year-over-year) and adjusted operating income to roughly $1.07 billion, also ahead of consensus estimates.
Risks remain, although they look different from the challenges most software companies face. Palantir’s international growth continues to lag the pace of expansion seen domestically, although its U.S. concentration arguably supports long-term predictability. Government contracts also remain an important part of the business, creating some exposure to budget cycles, procurement delays, and shifts in spending priorities.
At the same time, management has suggested that demand for its AI platform is arriving faster than PLTR can fully support in certain areas – not a typical problem for a growth company in the software space, rather resembling NVIDIA’s capacity constraints. Palantir’s main challenge is scaling its workforce, infrastructure, and deployment capabilities quickly enough to meet demand.
1
The Strategy Equation
Palantir occupies a unique position at the intersection of enterprise software, AI infrastructure, and mission-critical operations, making peer selection less straightforward than for a typical software company. Snowflake provides the closest comparison from a data-platform perspective, reflecting the growing importance of organizing and governing enterprise information in the AI era. CrowdStrike – a core Smart Investor Portfolio holding – offers a useful benchmark for premium software valuation, combining rapid growth, strong customer retention, expanding profitability, and an increasingly strategic role within enterprise and government environments. ServiceNow represents the scaled workflow and automation incumbent, while Datadog adds another perspective through its exposure to cloud operations, observability, and AI-driven software development.
That positioning also helps explain the wide divergence in stock performance across the peer group over the past year. Datadog has been the clear standout, with its triple-digit gain reflecting a powerful re-rating as investors increasingly viewed observability and infrastructure monitoring as direct beneficiaries of rising AI workloads. CrowdStrike also delivered strong returns, supported by resilient cybersecurity spending, expanding profitability, and growing recognition that AI increases the complexity of defending enterprise systems instead of reducing the need for security platforms. Snowflake posted a solid gain as sentiment improved toward data infrastructure, though investors continue to look for clearer evidence that AI adoption can translate into sustained acceleration. ServiceNow moved in the opposite direction, weighed down by concerns that agentic AI could eventually pressure portions of traditional workflow software and seat-based economics.
PLTR sits somewhere outside that framework. Its more modest gain does not reflect weaker business performance – quite the opposite, as revenue growth, margins, customer expansion, and cash generation all accelerated materially. Instead, much of the past year has been spent digesting a prior AI-driven re-rating while investors debated where Palantir belongs in the emerging AI stack. Increasingly, the answer appears to be neither model provider nor traditional software vendor, but the operational layer that allows organizations to deploy, govern, and execute AI inside real-world environments. As fears that frontier-model companies such as OpenAI and Anthropic could displace enterprise software have faded, the focus has shifted toward which companies help customers operationalize those models.
That shift increasingly appears to favor Palantir, and it’s becoming increasingly difficult to ignore in the valuation. Palantir still trades at a premium to most software companies, but the premium looks far less extreme once growth, profitability, and cash generation are all considered. Among its peer group, PLTR delivers the strongest revenue growth, the highest margins, the highest returns on capital, and the highest revenue per employee, while also generating substantial free cash flow. Yet its forward valuation multiples are no longer clear outliers. Palantir trades at roughly 109x forward non-GAAP earnings, below Snowflake, CrowdStrike, and Datadog, while its forward EV/EBITDA multiple is also lower than all three. ServiceNow remains the cheapest stock in the group, but it also carries a substantially slower growth profile, reflected in its stock’s underperformance.
The contrast becomes even more striking when viewed through a growth-adjusted lens. Palantir’s forward PEG ratio stands at roughly 2.1x, less than half Snowflake’s and well below CrowdStrike’s and Datadog’s despite materially faster expected growth. In other words, the market is still assigning PLTR a premium multiple, but increasingly appears to be paying for demonstrated execution, profitability, and growth instead of pure AI enthusiasm.
Analysts see an average upside of roughly 21% for the stock. However, the wide dispersion of price targets – from as low as $70 to as high as $255 – signals that Wall Street models are also struggling with the categorization of the stock, as pure software calls for lower valuations than infrastructure layer moats. Amid this wide disagreement about what Palantir fundamentally is, and what it is worth, we choose to stick with one of the most trusted voices among tech sector analysts, Wedbush’s Dan Ives, whose $230 price target implies roughly 50% potential upside. Ives called Palantir the “Messi of AI” and expects it to join the trillion-dollar market-cap club, as he sees its AIP platform as a unique, highly defensible software/AI layer. NVIDIA’s Jensen Huang has similarly argued that the rise of agentic AI makes this an “incredible time to be a software company,” reinforcing the view that value in AI may increasingly accrue to the platforms that operationalize and govern intelligent systems.
1
Investing Takeaway
Palantir has evolved from a niche government software provider into a foundational operating layer for the AI era. Its strength lies not in building the most powerful models, but in helping organizations deploy, govern, and act on them inside real-world environments where security, accountability, and execution matter. As AI adoption moves from experimentation toward operational deployment, Palantir’s position appears increasingly aligned with the next phase of the cycle. The company is expanding across enterprise software, defense modernization, sovereign AI, and industrial automation while deepening its role inside some of the world’s most important organizations. Execution risks remain, particularly as growth accelerates and deployment capacity is stretched. Still, few companies combine Palantir’s growth, profitability, strategic relevance, and ecosystem reach. Increasingly, it looks less like a software vendor and more like critical infrastructure for the age of operational AI.
1
New Sell: Labcorp Holdings (LH)
We are selling Labcorp Holdings because the stock has failed to respond to continued execution, leaving the position with limited near-term upside despite an intact business case. This is not a broken-company call. LH remains a high-quality healthcare infrastructure business with strong diagnostics scale, expanding specialty testing capabilities, solid central laboratory demand, and disciplined capital allocation. The problem is that the market still does not appear willing to reward those strengths.
The latest quarter supports that view. First-quarter revenue rose 5.8% year-over-year to $3.54 billion, adjusted EPS increased 10.6% to $4.25, adjusted operating margin expanded to 14.4%, and management raised full-year 2026 revenue and adjusted EPS guidance. Diagnostics revenue grew 5.0%, while Biopharma Laboratory Services revenue rose 8.2%, supported by 11.3% growth in Central Labs. Labcorp also continued to expand specialty diagnostics, health-system collaborations, AI-enabled operations, and consumer-facing testing tools.
In other words, the original thesis remains valid: Labcorp is still executing, still generating cash, still growing earnings, and still trading at a reasonable valuation. But the stock’s behavior is telling a different story. Since our purchase, LH has declined roughly 5%, even as the company delivered the kind of steady results that should have supported better performance.
The issue is attention. Labcorp is well covered by Wall Street, and institutional ownership remains high, but the story does not translate easily into the kind of market narrative that attracts broader investor demand. Its recent headlines are positive albeit incremental: cancer-testing collaborations, diagnostic launches, AI tools, health-system partnerships, and modest analyst target changes. These are real business developments, but they are not the type of news flow that moves today’s retail-driven tape.
Labcorp remains a strong company, but the stock is not being rewarded for that strength. With no clear catalyst for a near-term rerating, we prefer to exit and redeploy capital into names with stronger investor attention and a clearer path to upside.
1
Smart Investor’s Winners Club
The Winners Club represents stocks from the Smart Investor Portfolio that have risen at least 30% since their purchase dates.
Bulls have led the charge, and the Club member count expanded to 28: AVGO, GE, TSM, ANET, EME, HWM, APH, VRT, ORCL, IBKR, STRL, ASX, CRWD, MTZ, CSCO, GOOGL, PH, KEYS, PANW, JBL, ATI, BNY, MS, NVT, CRDO, RTX, and C.
The first runner-up is now SNPS with a 27.76% gain since purchase. Will it break into the winners’ circle, or will another stock outrun it to the finish line?
1
|
