Global investment in artificial intelligence infrastructure is projected to reach $1.3 trillion this year.. This massive capital expenditure is fueling unprecedented earnings growth for semiconductor companies while simultaneously pushing AI software providers like Palantir to extreme valuation multiples.

Advertisement

The $1.3 trillion migration toward semiconductor earnings

AI capital expenditure is expected to hit $1.3 trillion this year, a significant increase from the $1.1 trillion estimated earlier in the first quarter. As the report indicates, this surge in spending is driving a massive shift in free cash flow from hyperscalers directly into the semiconductor cohort.

Semiconductor manufacturers like Micron and Sandisk are anticipating massive jumps in earnings per share (EPS) as a result of this cycle . According to the analysis, Micron's EPS is projected to climb from $7 in 2025 to $150 by 2027, while Sandisk is expected to see its EPS rise from $70 to $250 in the same period.

Palantir’s 155 Rule of 40 score and the push for AI sovereignty

Palantir has emerged as a primary beneficiary of enterprise AI demand, converting foundation models into functional workflows through its Gotham, Foundry, and AIP platforms. The company reported revenue of $1.94 billion, which represents a 93% year-over-year increase.

The growth within Palantir is being driven by a dual focus on commercial and government sectors. u.S. commercial revenue surged 149% to $764 million, while U.S. government revenue grew 90% to $809 million. this performance resulted in an adjusted operating margin of 62% and a Rule of 40 score of 155, a level of hypergrowth and profitability that the report describes as having no precedent at this scale.

Snowflake’s $9.21 billion backlog and the AI Data Cloud pivot

Snowflake is transitioning its business model toward an "AI Data Cloud" to better capture the demand for optimized data estates.. The company reported product revenue of $1.334 billion for Q1 fiscal 2027,marking a 34% increase and its strongest sequential dollar growth in company history.

The company's scale is further evidenced by its $9.21 billion in remaining performance obligations. Snowflake has also seen its million-dollar customer count reach 779, with 46 of those customers crossing that specific threshold during the most recent quarter.

Will the 100x earnings multiple survive a cyclical downturn?

The current valuation of Palantir presents a significant risk for investors, as the stock trades at more than 100 times earnings. Because the stock is essentially a duration bet on hypergrowth persisting for years, any deceleration in U.S. commercial bookings could lead to a violent correction in the share price.

Analysts are still questioning whether the current demand for AI infrastructure is structural or merely a temporary cyclical peak. Furthermore, while companies like Nebius have secured contracts for A100 GPUs out to 2029, it remains unverified how quickly hyperscalers will successfully monetize their massive capital expenditures to sustain this level of semiconductor spending.