In the middle of the 19th century, Britain was swept up in what became known as the Railway Mania - a genuine investment frenzy in which everyone wanted to put money into railways. Hundreds of companies were formed, capital poured in torrents, share prices soared toward the sky, and the newspapers repeated that railways would change the world forever. Most of those companies, however, went bankrupt or accumulated enormous losses. They built lines that no one used sufficiently, spent without restraint, and never succeeded in turning the revolutionary technology into a sustainable business.
Today we are witnessing an almost identical picture with artificial intelligence. Billions are being directed toward training ever-larger models, toward computing power, and toward “tokens.” Almost everyone is investing in the “rails.” Yet just as 180 years ago, the true winners are the few who are not merely laying rails but running the trains - those who turn the technology into real operational value and profit. One of those rare companies is Palantir.
On 3 August 2026, Palantir released its results for the second quarter ended 30 June. The company once again smashed analysts’ expectations and raised its guidance for the third consecutive time. The quarter was described by management as “otherworldly,” and the growth rate was presented as unprecedented for a company of this scale.
Revenue - by Segment and Comparison with Analyst Expectations
Table 1. Palantir Revenue by Segment - Q2 2026 vs Analyst Expectations
| Metric | Q2 2026 | YoY Growth | QoQ Growth | Analyst Expectations |
|---|---|---|---|---|
| Total Revenue | $1.935 billion | +93% | +19% | ~$1.80-1.81 billion |
| U.S. Revenue | $1.573 billion | +115% | +23% | - |
| – U.S. Commercial | $764 million | +149% | +28% | - |
| – U.S. Government | $809 million | +90% | +18% | - |
| International (remainder) | ~$362 million | - | - | - |
For the first half of 2026, revenue reached approximately $3.57 billion. The company now generates more than 81% of its revenue from the United States (versus 73% a year earlier). The international business remains significantly smaller and is growing more slowly. This is not accidental: the adoption of operational artificial intelligence platforms outside the United States faces more serious regulatory, cultural, and competitive barriers. While in America “sovereign AI” is rapidly becoming a priority for corporations and agencies, in Europe and Asia the process is slower and more fragmented. This geographic concentration is both a source of strength at present and a structural vulnerability in the medium term.
Guidance for the third quarter of 2026: $2.160-2.164 billion.
New full-year 2026 guidance: $8.150-8.158 billion (+82% year-over-year). U.S. Commercial is expected to exceed $3.424 billion (at least +134% year-over-year).
Costs
- Cost of revenue: $297 million → gross margin of approximately 85%.
- Operating expenses: $727 million.
- Sales and Marketing: $340 million.
- Research and Development: $193 million.
- General and Administrative: $195 million.
- Stock-based compensation: $265 million (+66% year-over-year) + $17 million in employer taxes related to it.
Costs are growing significantly more slowly than revenue - a classic sign of scalability.
Profits
Table 2. Palantir Profitability Metrics - Q2 2026
| Metric | Q2 2026 | Margin | Year-over-Year |
|---|---|---|---|
| GAAP Operating Income | $912 million | 47% | strong growth |
| Adjusted Operating Income | $1.194 billion | 62% | - |
| GAAP Net Income | $1.062 billion | 55% | +225% |
| Adjusted Earnings Per Share | $0.41 | - | vs $0.34-0.35 expected |
| Rule of 40 | 155% | - | - |
Free Cash Flow - the Strongest Element in the Report
Table 3. Free Cash Flow - Q2 2026 vs Q2 2025
| Metric | Q2 2026 | Q2 2025 | Change | Margin |
|---|---|---|---|---|
| Cash from Operations | $1.216 billion | $539 million | +126% | 63% |
| Capital Expenditures | $14.6 million | $7.6 million | +91% | - |
| Adjusted Free Cash Flow | $1.220 billion | $569 million | +115% | 63% |
This is the first time Palantir has generated more than $1 billion in adjusted free cash flow in a single quarter.
How Adjusted Free Cash Flow is calculated: Cash from Operations + cash payments for employer taxes on stock-based compensation - purchases of property and equipment.
In Q2 2026, capital expenditures were symbolic ($14.6 million), which is typical for a software platform. The conversion of adjusted operating income into free cash flow is nearly 1:1.
Historical trend:
- Q2 2024: approximately $141 million (margin around 15%)
- Q2 2025: $569 million (57%)
- Q1 2026: $925 million (57%)
- Q2 2026: $1.220 billion (63%)
The company ended the quarter with $9.2 billion in cash and short-term U.S. Treasuries.
Full-year 2026 guidance: Adjusted Free Cash Flow has been raised to $4.5 - $4.7 billion (from the previous $4.2-4.4 billion).
This is one of the highest free cash flow margins in the public software sector and clearly demonstrates that the business is already operating as a classic “cash machine.”
New Customers and Segments
Table 4. Customers, Contracts and Growth Metrics - Q2 2026
| Metric | Value | Year-over-Year |
|---|---|---|
| Total customers | 1,049 | +24% |
| U.S. Commercial customers | 653 | +35% |
| Net Dollar Retention | 157% | - |
| Closed deals ≥ $1 million | 220 (of which 73 ≥ $10 million) | - |
| Total Contract Value | $3.373 billion | +49% |
| U.S. Commercial Total Contract Value | $2.132 billion | +153% |
| U.S. Commercial Remaining Deal Value | $6.238 billion | +124% |
Growth is coming both from new customers and (even more so) from expansion with existing ones, with a relatively small and efficient sales team. Net Dollar Retention of 157% is a particularly strong signal - existing customers are spending significantly more over time.
Competition
Palantir competes on several fronts simultaneously:
- Enterprise data and artificial intelligence platforms: Databricks, Snowflake, C3.ai, Microsoft Fabric, ServiceNow
- Large cloud providers: Azure AI Foundry, AWS Bedrock, Google Vertex
- Defense: Anduril and European alternatives (ChapsVision, Helsing, and others)
- Foundation model laboratories: OpenAI, Anthropic, and others
The company’s differentiation is based on the combination of ontology and operational artificial intelligence (AIP) - an approach that emphasizes real-world deployment and the customer’s full control over their data. This is a material difference from most foundation model companies, which sell access to models, and from pure data platforms, which focus on storage and analysis. The question, however, is how durable this difference is. Hyperscalers possess enormous distribution channels, embedded relationships with corporate customers, and nearly unlimited resources. If they succeed in offering sufficiently good “operational” layers on top of their own models, Palantir’s moat could be narrowed. At present the numbers show that this is still not happening at scale, but competitive pressure will intensify.
Valuation - the Price of Perfection
The stock trades at extremely high multiples - price-to-sales and forward price-to-earnings ratios in triple digits. The market has priced in an almost perfect scenario: continued hyper-growth, maintenance of margins above 60%, and successful expansion of the U.S. Commercial segment. To justify the current price, the company must meet or exceed its current forecasts for several years.
The flip side is clear. Any slowdown in growth below 50-60% annually, or compression of margins, or disappointment in the commercial segment, could trigger a significant re-rating. The history of highly valued software and technology companies is full of examples in which even good results proved insufficient if they were “less good” than what was priced in. High valuation does not make the company bad. It simply removes almost any margin of safety for the investor.
Red Flags, Risks, and Opportunities
Risks
High valuation (price-to-sales and forward price-to-earnings in triple digits)
The stock trades at extremely high multiples, meaning the market has already priced in an almost perfect scenario of continued hyper-growth. Any, even slight, slowdown in the pace could lead to a significant correction in the price, as expectations are exceptionally high.
Concentration: more than 80% of revenue comes from the United States
Such strong dependence on a single geographic market makes the company vulnerable to any changes in American policy, budget constraints, or economic conditions. If growth in the U.S. slows, the international business is still not large enough to compensate. Slower adoption outside America may prove to be a structural rather than a temporary factor.
Government contracts with termination-for-convenience risk
A large portion of government contracts contain clauses that allow the customer to terminate the contract at any time. This creates uncertainty in the long-term visibility of revenue, especially in the event of a change in administration or a review of defense and intelligence budgets.
Stock-based compensation continues to dilute shareholders
Stock-based compensation remains a significant expense and continues to increase the number of shares outstanding. Although adjusted metrics exclude it, the actual dilution reduces existing shareholders’ claim on future profits.
Competition from model laboratories and the major cloud providers
OpenAI, Anthropic, Microsoft, Amazon, and Google have enormous resources and are attempting to penetrate deeper into enterprise artificial intelligence. If they succeed in offering sufficiently good operational solutions, they could exert pressure on Palantir’s pricing and market share.
Opportunities
The trend toward “sovereign artificial intelligence”
More and more organizations and governments are insisting on retaining full control over their data and artificial intelligence models rather than handing them over to the large laboratories. This plays directly into the strengths of Palantir’s platform, which was designed precisely for this type of sovereign and secure deployment.
The U.S. Commercial segment still has significant room for expansion
Despite the impressive 149% growth, penetration of the corporate sector in the United States remains relatively low relative to the total addressable market. If the company continues to convert pilot projects into large-scale, high-value contracts, this segment can remain the primary growth engine for several more years.
Exceptionally high margins and cash flows
With an adjusted operating margin of 62% and a free cash flow margin of 63%, Palantir generates enormous amounts of free cash. This gives the company strategic freedom to invest aggressively, make acquisitions, or return capital to shareholders without relying on external financing.
Continuation of the current pace in the medium term
If growth is maintained at similar levels over the next 12-18 months, the company will continue to expand both revenue and profits at rates that few other public companies can match. The key question is whether this pace is sustainable after the most easily accessible opportunities have been exhausted.
Potential Through the End of 2026
The new guidance implies approximately $8.15 billion in revenue and nearly $4.9 billion in adjusted operating income, plus $4.5-4.7 billion in free cash flow. If the U.S. Commercial segment continues to grow at more than 130% and the free cash flow margin remains above 60%, the company will finish the year as one of the strongest cash-flow generators in the artificial intelligence and enterprise software space. If the current dynamics are maintained, Palantir has a realistic chance of continuing to positively surprise the market in the next two quarters as well. The opposite is equally true: any downward deviation from the projected trajectory will be punished sharply by the market because of the high valuation.
Conclusion
Just as during the British Railway Mania, most of the money invested in artificial intelligence today will likely turn out to be losing. Many companies will continue to burn capital in an attempt to build “longer and faster rails” without ever achieving sustainable returns. Palantir’s second quarter of 2026, however, showed something different: the company is no longer building rails - it is running the trains. With record growth, a 63% free cash flow margin, and a clear position in operational artificial intelligence, Palantir is among the few that are succeeding in turning the artificial intelligence revolution into real money.
The question is no longer whether the technology will change the world. The question is who will survive and profit when the mania fades - and at what price investors are prepared to pay for a seat on this train.
Frequently Asked Questions (FAQ)
1. What does the “Rule of 40” mean and why is it important for Palantir?
The Rule of 40 is a metric that adds the year-over-year revenue growth rate and the operating margin. A score above 40% is considered excellent for software companies. At Palantir this metric reached 155% in the second quarter, demonstrating an exceptionally rare combination of hyper-growth and high profitability.
2. What is the difference between GAAP metrics and adjusted metrics?
GAAP metrics include all expenses, including stock-based compensation. Adjusted metrics exclude stock-based compensation and the related taxes in order to show more clearly the operational profitability of the business. Investors monitor both, but adjusted figures are often used for comparisons between companies.
3. Why is free cash flow such a strong argument in favor of Palantir?
Free cash flow shows the real cash the company generates after all operating needs and capital expenditures. With a 63% margin and more than $1.2 billion in a single quarter alone, Palantir demonstrates that its growth is not merely “on paper” but is converting into genuine cash that can be used for investments, acquisitions, or capital returns.
4. What is “sovereign artificial intelligence” and why is it important for the company’s future?
Sovereign artificial intelligence means that organizations and governments retain full control over their data, models, and decisions rather than handing them over to external laboratories. Palantir is positioned precisely in this segment - it offers a platform that enables secure and controlled deployment of artificial intelligence, giving it a strong competitive advantage in the current market environment.
5. Can Palantir’s growth be sustained in the coming years?
Management has signaled that the current pace can be maintained in the medium term. The company raised its full-year 2026 guidance, and the U.S. Commercial segment still has significant room for expansion. Despite the stock’s high valuation, if growth and margins hold, Palantir has the potential to remain among the leaders in operational artificial intelligence. The key risk remains whether the current expectations priced into the stock will prove too ambitious.
Important notice
This article is for information and education only. It is not financial advice, an investment recommendation or professional advice of any kind. The views expressed are the author’s personal opinions and are not recommendations to buy, sell or hold securities, assets or investments. Investing involves substantial risk, including the possible loss of all capital. Past performance is not a guarantee of future results. Each reader should conduct independent research, consult a qualified financial adviser and weigh financial circumstances, risk tolerance and investment objectives before taking any decision. The author and borsite.net accept no liability for any loss arising from the use of the information in this article.