For much of the AI boom, winning has been synonymous with spending. Microsoft (MSFT), Alphabet (GOOGL), Meta Platforms (META) and Amazon.com (AMZN) are investing tens of billions of dollars in AI infrastructure, while OpenAI, Anthropic, Alphabet's Google and other developers continue racing to build the industry's most capable AI models.

Palantir Technologies (PLTR) chose a different path.

Rather than building its own AI model or investing heavily in data centers, the company built software designed to work with whichever AI model its customers choose. That means Palantir doesn't necessarily need one AI company to win. As businesses adopt AI across their operations, the company can potentially benefit regardless of whether the underlying technology comes from OpenAI, Anthropic, Google or another provider.

Growing Like Nvidia Without Building What Nvidia Builds

The strategy is translating into remarkable growth.

Palantir reported 93% year-over-year revenue growth in the second quarter, outpacing Nvidia Corp. (NVDA)'s 85% revenue growth reported in its most recent quarter. Over a longer period, Palantir has expanded from a $533 million quarterly business in the second quarter of 2023 to $1.94 billion in the latest quarter, growing its revenue base by more than 3.6 times in just three years.

Unlike Nvidia, however, Palantir isn't selling chips that power AI. Nor is it building the massive computing infrastructure required to train increasingly powerful AI models.

Instead, its software helps enterprises connect AI with their own data, workflows and day-to-day operations, regardless of which underlying AI model they use.

Billions in Cash Flow, Millions in Capital Spending

Palantir's business model also looks very different from much of the AI industry.

The company generated $2.1 billion in operating cash flow during the first half of 2026 while spending just $22 million on capital expenditures, or investments in long-term assets such as offices, equipment and infrastructure. That works out to well below 1% of revenue, illustrating how little physical infrastructure Palantir needs to grow compared with companies building AI data centers.

For perspective, that capex-to-revenue share stands at about 23% for Microsoft and Alphabet, 35% for Meta and 83% for Oracle Corp (ORCL).

The contrast underscores a broader shift taking shape across the AI economy. While one group of companies is investing enormous sums to build the technology that powers AI, another is focused on helping businesses actually use it.

Why Investors Should Watch

Palantir believes that second opportunity could prove just as important.

Shyam Sankar, Chief Technology Officer argued during the earnings call that the industry has become too focused on creating more powerful AI while paying less attention to turning that technology into measurable business results.

"The market has created far more intelligence than it has converted into value," Sankar said.

For investors, that may be the bigger story. Rather than betting on a single AI winner, Palantir has positioned itself to benefit from the broader adoption of AI across enterprises—regardless of which company ultimately builds the dominant model.