Palantir is a company that inspires "roller coaster" reactions among investors, and the current picture is a study in contrasts. On one side, the underlying demand for what the company does is exceptionally strong. On the other, Wall Street sentiment has cooled significantly. It is precisely this discrepancy — robust real-world demand paired with skeptical investor sentiment — that makes the stock attractive from a value-and-signal perspective.
The Core Debate: Is It Just a Software Company?
At the heart of the discussion is a question: is Palantir essentially a software company, and if so, how vulnerable is it to commoditization as AI advances? The concern is that as artificial intelligence continues to improve, the tasks Palantir performs could eventually be replicated more cheaply and easily, eroding its value. This is the central long-term risk, and it is worth examining in detail — but the counterargument is that Palantir is meaningfully different from a pure software business.
Business Momentum and Recent Milestones
The company's operating trajectory is impressive. Revenue over the most recent quarter grew somewhere in the range of 70–85% on a year-over-year basis, with the growth rate cited at roughly 85%. The number of customers rose by 31%. Despite that explosive growth, total revenue remains relatively modest at approximately $1.6–$1.8 billion, which is part of why the valuation debate is so heated.
Several concrete achievements underline the breadth of Palantir's reach:
- Healthcare breakthrough: A collaboration with Tampa General Hospital cut sepsis deaths in half across a cohort of 900 patients.
- Military integration: Palantir is embedded with the U.S. military, and the Army's next-generation command and control center runs on Foundry.
- Partnerships and international expansion: The company has partnerships with NVIDIA, and Japan is moving toward adopting its technology for military use.
What makes the offering compelling is the core function: Palantir turns messy, unstructured data into decisions. But it is far more than a simple software layer. Once a customer is onboarded and begins using the systems, Palantir's own on-site engineers work to ensure everything runs smoothly. This deep integration means the product becomes woven into the fabric of how the customer operates. Customers — especially government clients bound by long-term contracts — simply do not walk away. These are durable, sticky relationships.
Valuation: Paying a Premium for Growth
The valuation is undeniably rich, though the exact figure depends on which numbers you cite. The forward price-to-earnings ratio could be somewhere in the 80s or the 50s — roughly 75–80 times next year's earnings on some measures. For comparison, a company like Microsoft trades in the 30s. Buying Palantir therefore means paying a premium for anticipated future growth.
However, there is a mathematical logic that softens the concern. When you sell the best software and steadily roll customers into ever-larger deployments — with revenue and earnings potentially doubling every year — it does not take long for lofty valuations to compress on their own, because the earnings and revenue base grows so quickly beneath the price. Combine that dynamic with roughly 80% margins and large contracts locked in with customers who will never leave, and the business model becomes genuinely attractive despite the elevated multiple.
The Sentiment Data: A Large Mispricing
The demand-side data tells a clear story. On a scale of 100, Palantir scores an 87 on consumer demand — a high-quality score. This measurement draws not only from social media activity but also from website visits, which can capture enterprise-side inquiries and similar signals of genuine business interest.
The key insight comes from comparing two numbers: the consumer demand score against the investor sentiment score. The gap between them is where mispricing is believed to live. Palantir shows one of the largest such gaps — a difference of about 57 points — making it one of the most mispriced names tracked. By contrast, a company like Snowflake scores 72 on consumer demand, but investors largely agree with that reading, so there is little edge there. Big Bear shows a large discrepancy as well, but Palantir's is the widest.
The investor sentiment side appears to be driven largely by what the stock price itself has done, which has been highly volatile: shares were at $207 not long ago, then fell to $106, and now sit around $127. That kind of price action feeds a negative sentiment loop even as the underlying business strengthens.
Looking at the chart, the blue lines represent consumer demand data (including enterprise data), while the dotted line is the stock price. The gap is not enormous, but the big run in consumer data did not get the follow-through from the stock price that one would expect — despite revenues rising, earnings rising, and everything that should have been pushed up by that demand actually being pushed up. In other words, the fundamentals moved, but the price did not fully follow.
The Moat: Why Palantir Is Not Just Another Software Name
Question: Is the moat — particularly the room to grow on the commercial side — what makes the data so exciting, given that the total addressable market isn't close to being monetized yet?
The total addressable market is massive. It is essentially any large company burdened with a ton of unstructured, messy, or legacy data systems — and that describes almost every large organization. Unless a company was founded in the last five to ten years or is deeply immersed in the tech space, it is almost certainly sitting on ugly legacy data systems that need to inform decisions. The fact that Palantir can serve clients ranging from the military to hospitals implies it can serve the vast universe of companies in between. This enormous, largely un-monetized market is a central reason for optimism.
The deeper argument, though, is about defensibility. There is a real risk that AI development commoditizes software broadly — many worry Palantir has been unfairly thrown in with the rest of the software sector. But Palantir is likely to be one of the last software companies to be killed by AI, precisely because its moat is so much larger. Companies like Salesforce, HubSpot, Dropbox, and possibly even Adobe are far more replicable — they are relatively easy to clone, after which success is mostly a matter of sales. Their systems would fall to AI-driven competition long before Palantir's.
Palantir is different because of how deeply it integrates:
- It uses air-gapped technology, with servers sitting physically inside the client's facility.
- It places engineers on-site, working directly alongside the customer.
- The overall engagement is far more involved than software alone.
This makes Palantir extraordinarily difficult to clone. If one were building a software company to displace enterprise legacy systems, Palantir would not be a target — one would pick far easier prey first.
That said, the long-term risk is real and should not be dismissed. AI is advancing so quickly that comparing its capabilities today with a year ago is astonishing; extrapolating five years forward is almost impossible to imagine. AI may eventually be able to do what Foundry does. If so, Palantir has perhaps five years — or whatever that window turns out to be — to embed itself even more deeply and make its product even stronger. It may well survive the transition, but the threat is genuine over a long enough horizon.
Sentiment, Not Fundamentals, Driving the Discount
Question: Palantir was once even more expensive than it is now, and investors kept piling in — so is the current weakness partly just a matter of being caught on the wrong side of investor sentiment?
Yes, largely. There is an element of "throwing out the baby with the bathwater." The underlying data on the company is strong, and while the AI-development concern is legitimate, Palantir is simply different in what it does. The sell-off therefore looks somewhat unwarranted. Because Palantir got lumped in with the broader software trade — amid rotations out of specific software names — it has been swept up on the wrong side of sentiment even though its fundamentals justify better treatment.
The Verdict: A Medium-Term Opportunity
Weighing everything together, the conclusion is that this is more of an opportunity than a warning. The judgment on the very long term is uncertain — the AI risk keeps that horizon open. But for the medium term, roughly a one-to-two-year or even one-to-three-year hold period, Palantir looks like a strong opportunity. The sell-off appears unjustified relative to the demand data, the business is compounding rapidly, the margins are exceptional, the contracts are effectively permanent, and the moat is uniquely difficult to replicate. In short, investors appear to be discarding a strong company along with the weaker software names, and that creates an attractive entry for a one-to-three-year hold.
