Applied Digital Delivers Polaris Phase, but Insider Selling Casts Shadow on AI Infrastructure Play
Published on 07/20/2026 at 17:14 | Redaktion boerse-global.de
Applied Digital marked a key construction milestone on Monday as it delivered the first phase of its second building at the Polaris Forge 1 campus, adding 75 megawatts of capacity and bringing the total online to 175 MW. The facility, designed for high-density AI and HPC workloads, is set to reach 400 MW of critical IT load at full buildout, underpinned by long-term leases that support the company’s so-called “AI Factory Franchise Model.” The on-time delivery in Texas is seen as a test of whether Applied Digital can replicate that blueprint at other sites – a linchpin of its growth narrative.
The market gave the development a tepid nod, with shares rising 7.59% on the day to €23.82. That bounce did little to erase the pain of recent weeks, however. Over the prior 30 days, the stock had shed roughly 41% of its value, and at Friday’s close of €22.14 it stood nearly half below the 52-week high of €43.48 set in late May. The relative-strength index, which dipped to 26.2 on Friday before recovering to 32.8 after Monday’s gain, signals deeply oversold conditions – a technical cue that the sell-off may have overshot. Annualized volatility of 74% underscores the jittery tone around the name.
Adding to the tension, insider selling has accelerated even as institutional investors have quietly added to positions. Over the past 90 days, insiders sold 35,000 shares worth $1.18 million, and five transactions in July alone totaled $10.89 million. In a contrasting move, Sigma Planning Corp bought 45,858 shares in the first quarter, worth about $1.09 million. Large holders include Hood River Capital with 23 million shares (8.07% of outstanding), BlackRock at 7.31%, Vanguard at 5.18%, and management itself holding roughly 9.35%. The divergence between institutional accumulation and insider profit-taking is likely to keep investors on edge.
Should investors sell immediately? Or is it worth buying Applied Digital?
Applied Digital is posting explosive revenue growth – $126.6 million in the fiscal third quarter, up 139% year over year – but the economics remain punishing. A net margin of minus 78% and total debt of $2.6 billion highlight the capital-intensive nature of the business. At a price-to-sales ratio of 21, the stock trades at more than three times the valuation of rival CoreWeave, which commands a P/S of 6 despite carrying $25 billion in debt and generating over $2 billion in quarterly revenue. CoreWeave also posted a $740 million net loss in the first quarter, alongside a $99.4 billion backlog. The comparison shows that Applied Digital is being rewarded for its growth trajectory, but the scope for disappointment is correspondingly large.
All eyes now turn to the fiscal fourth-quarter report, scheduled for release on July 27 after the closing bell, followed by a conference call with management. Analysts remain broadly optimistic, with a mean price target of $70.56 and individual targets around $54 implying roughly 86% upside from current levels. For a stock that has lost nearly half its value in two months despite an operational milestone, the earnings call will be a critical test. Investors will be looking for convincing signs that revenue momentum can translate into improving margins – and that the operational engine underpinning the AI factory model is truly scalable. If the numbers deliver, the oversold condition could spark a sharp reversal. If not, the gap to the 52-week high may persist.
Ad
Applied Digital Stock: New Analysis - 20 July
Fresh Applied Digital information released. What's the impact for investors? Our latest independent report examines recent figures and market trends.
Disclaimer regarding our articles: No investment advice, no buy or sell recommendation. Information on prices, companies, and markets is provided without guarantee; changes are possible at any time. Stock market transactions can lead to substantial losses. Our articles are created and reviewed in whole or in part automatically with the support of AI.
