If your feed has been full of shaky stock charts and hot takes about AI “agents,” you’re not sleeping in your dreams. In early February, software stocks around the world sold off hard after new AI features showed they can do entire tasks end‑to‑end, not just chat in a sidebar like ChatGPT. Global market wraps tied the move to an AI‑led reset in expectations for enterprise software. In particular, coverage highlighted how Anthropic’s update to its “Cowork” tools spooked investors by automating work that many software products monetize per human user. [cnbc.com] [marketplace.org]

Some outlets even put the single‑day damage near $300 billion as investors re‑priced what SaaS is worth in an “agentic” world, where a digital helper can log in, read files, and finish jobs. And the weakness didn’t end that day; roundups noted steep year‑to‑date declines across big names like Adobe, Salesforce, and ServiceNow as the month wore on.[economicstime.com] [fool.com]

The through‑line: the market tried to price, very quickly, what happens when the definition of a “user” changes.

For two decades, SaaS growth rode a straightforward logic: more people using the tool = more licenses sold. That’s per‑seat pricing in a nutshell. Now bring in AI agents, it’s software that can plan steps, call tools, and complete tasks under your supervision. If a single agent lets one employee do what used to require five, a company may buy fewer seats. That fear even got a name in analyst notes: seat compression. Salesforce often shows up in these conversations as the “classic CRM example,” where investors worry that seats stop growing at the same pace once agents automate routine pipeline work. [fool.com]

None of this means “software is over.” It shows how value is measured and billed, was tied to human headcount, and agents broke that clean correlation. You can feel the anxiety in the way the sell‑off spread across regions and categories in those first days. [cnbc.com]

Plenty of smart people pushed back on the doom narrative. Goldman Sachs’ David Solomon called the reaction “too broad,” arguing the blanket panic didn’t match the nuance of how AI will be adopted. JPMorgan research went further, calling the logic “broken”: markets can’t simultaneously believe AI will decimate software demand and that AI infrastructure spending won’t pay off—those two beliefs cancel each other out. [fool.com] [fool.com]

A more grounded take: this is a reset, not an extinction event. Software that adapts by embedding agents and rethinking pricing will stabilize; software that can’t show clear outcomes will feel more pressure. That’s why February felt so volatile, stocks move faster than roadmaps, and it takes time for pricing and packaging to catch up. [fool.com]

The clean story is that we’re shifting from pure per‑seat to hybrid pricing. Expect combinations of:

Bain’s 2026 brief says per‑seat isn’t “dead,” but new models are gaining steam, especially as AI introduces background execution that doesn’t map to a login. McKinsey reaches a similar conclusion: with AI turning software into an active performer (not just a tool), vendors need new telemetry and go‑to‑market motions to capture value beyond headcount. Hybrids are the practical bridge for 2026–2028. [bain.com] [mckinsey.com]

Zoom out and the trend is bigger than pricing. Deloitte’s 2026 predictions argue that as AI agents spread across SaaS, products will look more like orchestrated services that learn and act across workflows—another reason the old “user = value” shortcut keeps breaking. [deloitte.com]

This is your sign to shift from listing tool names to showing outcomes. A short line like “Automated weekly report prep (summarize PDFs → update Excel tracker → draft Slack recap), saving 3 hrs/week” proves two things:

Before renewal season, run a quick seat audit. Who logs in? What’s shelfware? Which tasks are already agent‑friendly? Use that baseline to negotiate usage/outcome pilots alongside your core seats. Pricing advisors and operator briefs suggest that hybrids are the near‑term reality and that vendors are more open to experimentation during this transition. If underused licenses have ever burned you, this is your moment: procurement conversations are shifting your way. (Plenty of buyer guides are already coaching teams to cut seat waste and align contracts with actual consumption.) [bain.com] [saas-review-hub.contentwave.net]

Listen to earnings calls for three words: “seats,” “agents,” “pricing.” They’re breadcrumbs. When a vendor starts talking openly about agent attach rates or pilots for usage/outcome tiers, you’re hearing how fast customers are moving. S

The February slump wasn’t just a market mood swing; it was the first loud, public argument about how we measure software’s value when work is increasingly done by agents, not only humans. The old per-seat still matters, but it can’t carry the story alone. Expect hybrid pricing to be the default bridge as vendors prove outcomes and buyers insist on paying for what actually gets done. [bain.com], [mckinsey.com]

If you’re a student, build a tiny agent workflow and put the outcome on your resume. If you’re buying, run the seat audit and pilot usage/outcome terms. If you’re building, measure what your agent really changes—and price to match. The “SaaSpocalypse” isn’t the end of software; it’s a nudge to rethink the metrics in an agentic era. [deloitte.com]

Bain & Company. (2026). Per-seat pricing isn’t dead, but new models are gaining steam. https://www.bain.com/insights/per-seat-software-pricing-isnt-dead-but-new-models-are-gaining-steam/

Deloitte. (2026). TMT predictions 2026: AI agents reshaping SaaS models and pricing. https://www.deloitte.com/us/en/insights/industry/technology/technology-media-and-telecom-predictions/2026/saas-ai-agents.html

Economic Times. (2026, February 9). Why Anthropic’s Claude CoWork sparked $300 billion tech market sell-off this month: See which sectors are winning and losing amid AI boom. https://economictimes.indiatimes.com/news/international/us/why-anthropics-claude-cowork-sparked-300-billion-tech-market-sell-off-this-month-see-which-sectors-are-winning-and-losing-amid-ai-boom/articleshow/128122187.cms

The Motley Fool. (2026, February 6). Analyst “seat compression” lens: Should you buy Salesforce stock before Feb? https://www.fool.com/investing/2026/02/06/should-you-buy-salesforce-stock-before-feb/

The Motley Fool. (2026, February 16). Goldman Sachs CEO David Solomon says the tech sell-off is “too broad.” https://www.fool.com/investing/2026/02/16/goldman-sachs-ceo-david-solomon-says-the-tech-sell/

The Motley Fool. (2026, February 17). JPMorgan research says “broken logic” is driving the software sell-off. https://www.fool.com/investing/2026/02/17/jp-morgan-research-says-broken-logic-is-driving-th/

The Motley Fool. (2026, February 18). The 2026 software stock sell-off: 25–30% YTD declines across major software names. https://www.fool.com/investing/2026/02/18/the-2026-software-stock-sell-off-ai-disruption-fea/

McKinsey & Company. (2025). Upgrading software business models to thrive in the AI era. https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/upgrading-software-business-models-to-thrive-in-the-ai-era

Miao, H. (2026, February 4). Software stocks plunge as AI disruption fears grow. CNBC. https://www.cnbc.com/2026/02/04/software-stocks-plunge-us-ai-disruption.html

Ryssdal, K., & Bond, D. (2026, February 6). Why Anthropic’s update spooked software investors. Marketplace (APM). https://www.marketplace.org/story/2026/02/06/why-did-anthropics-ai-spark-a-stock-sell-off

TS2 Space. (2026, February 25). Salesforce stock slides as AI selloff bites software again; Feb 25 earnings now in focus. https://ts2.tech/en/salesforce-stock-slides-as-ai-selloff-bites-software-again-feb-25-earnings-now-in-focus/