Software beats hardware: Palantir surges on AI deal while semiconductor stocks falter

Verdict: The rotation thesis is real, but the trade is still a watch-list idea — wait

The AI-rotation thesis has genuine fundamental fuel: Palantir grew revenue 70.0% year over year in fiscal 2025 to $4.5B, posted a 54.9% net margin and $1.2B of free cash flow in the quarter ended June 30, 2026, and jumped about 9% on the NVIDIA sovereign-AI deal per the Yahoo Finance report. But two things keep this on the shelf today. First, the setup never fired once across 60 months of daily bars, no robust parameter setup was established, and live conditions do not match: PLTR itself is down 4.5% on the day, so the required divergence (chips down while Palantir closes green) is absent, and trend strength at 4.2 sits far below the 20 threshold. Second, the filings and fundamentals lean against the thesis — the June 30, 2026 ownership filings show net open-market selling of roughly $136.6M at Palantir and $27.0M at Palo Alto Networks, and Palo Alto's quarter ended April 30, 2026 swung to a $177M net loss with an operating margin around -6.1% — a delayed-in-coverage filing, but a directional counterweight to the "smart money rotating in" story. A follow-up ownership filing showing insiders still selling heavily, or a Palantir close below its 30-day moving average at $164.55 after entry, would flip this verdict.

Trade now: PLTR is one condition away — ADX is the gate

## Where the setup stands PLTR closed at $174.33, down 4.5% on the day, and the divergence premise is half-alive: the one-day move of -4.5% is well past the entry requirement of below -1.5%, the price trend is positive, and the 20-day moving average ($172.60) sits above the 50-day average ($158.30). What is missing is trend strength — the 14-day trend-strength reading is 4.2 and needs to be above 20. That is the only gate between you and a trigger, and it is roughly 16 points away. The rules were evaluated on real daily bars and simply have not opened an entry yet; this is a watch-list setup, not an active signal. ## What a triggered entry would look like The entry also requires the day's low to touch the nearest support level ($176.50 area per the live level map) while the close holds above it — a dip-and-reclaim condition that fits the thesis's "shakeout then rotation" logic. If triggered, the mechanical risk frame is tight by design: a fixed stop at -2.4% on the position and a take-profit at +4.8%, an effective reward-to-risk of about 2:1, with a hard signal exit if PLTR closes below its 30-day moving average, currently $164.55, and a 30-day maximum hold. ## What "wait" means concretely Do nothing today. Watch for a session where PLTR's one-day move is below -1.5% while trend strength climbs back above 20 — historically, strength rising through 20 marks the start of a directional leg, which would mean the selloff is resolving into a trend the strategy can ride. Note the tactical tension: the entry wants a sharp red day, so a calm drift higher delays the trigger rather than invalidating the thesis. Size any eventual fill at a fixed 2.4% account risk per the strategy's sizing rule, and remember PLTR's own history — a maximum drawdown of 48.2% over the lookback — argues for respecting that stop without exception.

Why the bull case still has support

The core of the idea — that AI money is rotating from hardware to software — is visible in the most recent fundamentals. Palantir is the cleanest expression of the thesis: for its fiscal year ended December 31, 2025, revenue grew 70.0% year over year to $4.5B, gross margin reached 84.6% (the 92nd percentile of 771 information-technology peers), and operating margin hit 41% (the 98th percentile of 861 peers). The momentum did not stop at fiscal year-end: in the quarter ended June 30, 2026, revenue rose another 18.6% sequentially to $1.9B, net income grew 22% to $1.1B, and free cash flow climbed 35% to $1.2B. Per the Yahoo Finance report, Palantir jumped about 9% on the NVIDIA sovereign-AI deal — a software name capturing the AI value chain while chips wobble, which is precisely what the idea argues. Palo Alto Networks supplies the second leg. Even setting aside its weakest recent quarter, the trailing fiscal year (ended July 31, 2025) shows free cash flow of $934.8M — the 96.6th percentile of 800 sector peers — on a 73.2% gross margin, and the company returned to growth in the quarter ended April 30, 2026 with revenue up 15.7% sequentially to $3.0B and operating cash flow up 57.2% to $871M. The cited market action supports the rotation narrative too: per CNBC, the Nasdaq fell as chipmakers struggled on July 2, 2026, the same stretch in which Palo Alto climbed roughly 4% per the Yahoo Finance piece. Two software names outperforming a falling chip index is the divergence the thesis is built on. The balance-sheet picture backs the durability argument. Palantir's debt-to-equity ratio fell to zero by the end of 2021 and stays at zero through the latest period, and its net margin has expanded relentlessly — from 16.6% in Q1 2024 to 24.2%, 32.6%, 43.3%, 53.3%, and now 54.9% across the last five quarters. These are the operating-leverage characteristics of a company that converts AI demand into profit, not just hype. On the strategy itself: the entry conditions — a semiconductor index down more than 1.5% in a day while Palantir closes green…

Scores

• Conviction score breakdown: 46
• Thesis support: 62
• Trade readiness: 30
• Risk quality: 45
• Trigger proximity: 35
• Fundamentals trend: 58

Watch items

• PLTR — ADX (14)
• PLTR — ROC (1) on a red day
• PLTR — Price vs 30-day moving average
• PLTR — Insider net open-market flow
• PLTR — Quarterly revenue growth
• SOXX — Single-day decline