Nicholas Mugalli
Institutional TMT & Macro Research, AI/Semiconductors/Policy
Legacy SaaS names such as Salesforce, ServiceNow and Adobe are losing their high-margin moat as they pay hyperscalers for non-discretionary AI capability. Median gross margins across major software players have bled lower since late 2023, with the bottom quartile down nearly 2.5% by 1Q26. As seat-based models break down, application software absorbs higher compute, model and infrastructure costs just to stay competitive, so the compression is no longer cyclical noise. x.comÂ
Usage-based and credit-burn pricing is driving large revenue uplifts at Adobe, Microsoft, Salesforce and ServiceNow, as multi-step agentic workflows consume tokens at exponential rates. Market value is shifting to software vendors that can map compute usage to real business outcomes, reviving subscription models written off as dead legacy. x.comÂ
ServiceNow's slowing contracted-revenue growth suggests agentic AI and growing open-source alternatives may be reducing the need for its products. Q3 operating margin fell to 9.4% from 16.8% a year earlier, net income dropped, and cRPO growth slowed from 25% to 20.5%; Q4 guidance implies 18.5%. x.comÂ
Alphabet's AI capex payback is being repriced, making its post-earnings decline more than a one-day mood swing. Guidance and cloud and AI margin commentary are the near-term tests. If growth continues but the multiple compresses, the risk is duration in mega-cap AI, not a collapse in technology growth. x.comÂ
Software stocks such as Adobe sold off on a threat that was never real: the AI labs' internal plans to compete with them were delusional and never a real possibility. x.comÂ
Accenture and Wolters Kluwer benefit massively from cheaper AI tokens and margin gains as they deploy AI. Their knowledge costs rank among the highest relative to COGS, but regulatory capture, reputation and sticky customers keep revenue broadly intact. Neither shows clear signs of slowing growth; Accenture guided to high-single-digit revenue growth from 2029. Despite enormous terminal-value compression, an unwinding AI-infrastructure trade could bring a sizeable rerating if fundamentals hold. … Show more
AI data-center investment needs outside AI spending to rise 33-fold by 2030 to break even. Hyperscalers hold $4.5 trillion in debt and future liabilities, but corporate and SaaS token spending outside circular deals is only $60 billion annualized against a $2 trillion break-even ARR. SK hynix says memory shortages may last until 2031, raising capex and CoreWeave default risk; Mitsubishi warns turbines are booked through 2031 and cost 400% more, delaying CoreWeave and Oracle data centers. x.com … Show more
Software looks like a group that can lead the market higher, with many names from Palantir and ServiceNow to Zeta and MongoDB moving from late Stage 4 or Stage 1 into early Stage 2 uptrends. Sentiment is at an all-time low on fears that AI agents and chatbots will replace SaaS. AI will disrupt some areas, but many software businesses are using it to grow revenue, ship faster and expand operating leverage, and will thrive. The charts will separate survivors from the dead. x.com x.comÂ
AI is forcing a repricing of the enterprise software business model, and for ServiceNow and SAP the reset is not finished. Claude Managed Agents eased production AI deployment, ServiceNow's AI bundling raised monetization questions, and SAP warned of short-term pain like its cloud-migration margin squeeze. Multiples compress before new economics are understood, and most software still trades at a premium on GAAP P/E and EV/FCF ex-stock comp. Workday at 1.7x revenue hints at a floor. x.comÂ
Software companies face a multi-year repricing because AI has removed the scarcity that made software valuable. For 20 years building software required engineers, letting SaaS firms charge monthly fees forever; when anyone can build a working product in a weekend with Claude, software stops being the moat. Atlassian is down 75%, HubSpot 69% and Figma 86%, and layoffs show headcount no longer scales with output. Survivors will be those with distribution, data and trust underneath. x.comÂ
Enterprise software names such as Salesforce, GitLab and Atlassian now trade at roughly replacement cost and should recover about half of what they lost, the usual pattern after a panic. Valuing them on replacement cost rather than profit makes them look different: the bulk of AI value is enterprise-facing, and entrenching systems inside enterprises takes so long that, with people talking about a singularity within two years, no one will have time to build competitors. x.com x.comÂ
AI usage outstripping compute supply for the foreseeable future supports semiconductor demand despite cheaper tokens. Token prices fell about 90% in roughly three months, yet usage rose by more than that drop, so compute pricing keeps rising. AI revenue extends beyond LLM providers, including Meta, Google, Salesforce and ServiceNow. On- and off-balance-sheet debt is a risk; an AI bubble is likely eventually, but none is visible now. x.comÂ
