AlixPartners scored 500 software companies for AI vulnerability last week. The results read like a triage report.
The consulting firm — which flagged AI’s threat to software a year ago, before most investors took it seriously — developed an “AI Disruption Score” ranging from 1 (insulated) to 7 (structurally exposed). They examined companies across 12 private-equity portfolios and assessed them on two dimensions: proprietary data depth and vertical specialization. Only 14% of companies scored well on both. Roughly a quarter had weak defenses on both fronts.
The pattern is specific enough to be useful. Marketing automation, horizontal productivity tools, CRM add-ons, analytics dashboards — anything where AI can replicate the core feature set with low switching costs — clusters at the high-disruption end. Payments infrastructure, healthcare systems, cybersecurity, financial operations software — anything embedded in regulated, mission-critical workflows with years of accumulated proprietary data — sits at the low end.
“If you’re a large financial institution, fraud detection software is mission-critical, and it’s not something you have any tolerance for error on,” Jordan Berger, AlixPartners SVP, told Business Insider. “These companies are much less likely to let agentic AI native challengers run rampant throughout their enterprise ecosystem.”
The interesting finding isn’t the extremes — everyone knew CRM add-ons were more vulnerable than compliance software. It’s the middle. Systems of record like ERP, CRM, and IT service management platforms rate only medium-strength moats. Even these incumbents face seat compression as AI agents reduce the number of humans who need licenses, and AI strips away the higher-margin add-ons and interfaces that historically padded revenue.
The debt wall nobody’s talking about
Here’s where AlixPartners’ analysis turns from strategic to financial. PE firms acquired SaaS companies in bulk during the low-rate era, drawn by the predictability of subscription revenue. Those acquisitions were debt-funded. There is now a $40 billion wall of software debt maturing in 2028 that will need to be refinanced — right as AI disruption hits the revenue lines backing that debt.
AlixPartners expects lenders to add roughly 50 basis points to refinancing rates. That’s the charitable scenario. If lenders priced the actual AI disruption risk into rates — roughly 400 basis points higher — “then it’s over,” as Nenad Milicevic, an AlixPartners partner, put it. “And that has a contagion effect on the whole market.”
The firm projects SaaS revenues could decline up to 15% over the next year and 25% to 35% over three years in the most exposed segments. Against that backdrop, they sort the universe into four categories: “fortress” businesses with strong moats, “survivors” that must acquire AI capabilities fast, companies likely to be sold to AI-native buyers, and those facing potential wind-down.
The counter-evidence: ServiceNow says hello
While AlixPartners maps which companies will get sorted out, ServiceNow is busy proving that the survivors can thrive. Its Now Assist generative AI suite surpassed $600 million in annual contract value in Q4 2025, more than doubling year over year, and is tracking toward $1 billion ACV in 2026. Enterprise customers are paying a 25% to 45% premium over standard tiers for AI-powered capabilities. Deals over $1 million in ACV nearly tripled quarter over quarter.
The S&P 500 Software Index has recovered nearly 15% from its February lows. ServiceNow’s stock surged 5.5% on April 1. But here’s the tension that makes this genuinely complicated: even ServiceNow is down 33% year-to-date. BTIG just cut its price target to $185 from $200. Goldman Sachs trimmed to $188. Stifel cut to $135.
ServiceNow’s FY26 subscription revenue guidance projects 19.5% to 20% growth — but that includes roughly 100 basis points from the Moveworks acquisition. The organic picture is solid but decelerating. BTIG’s concern isn’t that the business is broken; it’s that consensus estimates for FY27 and FY28 are too optimistic.
So ServiceNow is simultaneously the best evidence for AI-driven software adaptation and a company whose own analysts keep cutting their targets. It sits in AlixPartners’ “fortress” category — system of record, deep data moat, workflow gravity — but even fortresses trade at a discount to where they were six months ago.
What this tells us about the hypothesis
The hypothesis says software loses its value because AI commoditizes it. AlixPartners’ data tests this with unusual precision. Their framework confirms that commoditization risk is real and measurable — a quarter of PE-backed software companies have weak defenses against it. But it also shows that the disruption is radically uneven. The 14% with strong moats aren’t just surviving; they’re the ones building the AI capabilities that threaten the other 86%.
The debt wall introduces a mechanism the hypothesis didn’t originally contemplate: credit contagion. AI doesn’t just compress software valuations in the stock market. It potentially triggers a credit event in private markets, where $40 billion in debt sits against revenue lines that AlixPartners says could shrink by a quarter over three years. The sorting isn’t hypothetical — it has a deadline.
The falsification evidence is real too. ServiceNow’s $600M ACV for Now Assist demonstrates that system-of-record incumbents can successfully monetize AI at premium pricing, expanding revenue per customer even as seat counts face pressure. The 15% index recovery suggests the market is beginning to differentiate between categories rather than selling software wholesale.
The hypothesis needs a qualifier: software doesn’t uniformly lose its value. It bifurcates. The fortress companies that own proprietary data, sit in regulated workflows, and can charge for AI outcomes retain or grow their value. Everything else faces a reckoning — and the $40 billion debt wall gives that reckoning a date on the calendar.
Sources: AlixPartners via Business Insider, ServiceNow Q4 FY2025 earnings, BTIG analyst note via 24/7 Wall St., FinancialContent market analysis