Anthropic Claude “SaaSpocalypse” selloff wipes out $300B in software value

The “SaaSpocalypse” selloff is the first major market event explicitly blamed on agentic AI tools moving up the value stack. Anthropic’s new Claude Cowork plugins – especially the legal workflow agent – crystallized a fear that has been simmering for two years: if a $20-per-seat AI agent can navigate Salesforce, review NDAs, reconcile data, and write reports, what happens to software license margins and labour-intensive IT services? Indian IT names, BPOs, and Western SaaS giants sold off in unison as investors repriced the durability of business models built on billable hours and per-seat software.([economictimes.indiatimes.com](https://economictimes.indiatimes.com/markets/stocks/news/rs-1-75-lakh-crore-saaspocalypse-for-it-stocks-explained-what-it-means-for-investors/articleshow/127900151.cms?from=mdr))

For the race to AGI, the signal is profound. Markets are starting to treat certain kinds of “boring” enterprise cognition – compliance checks, migrations, testing, reporting – as within reach of near-term AI agents. That implicitly raises expectations about the pace of capability growth and about where in the software stack value will accrete: orchestration, data moats, and governance, rather than raw APIs or headcount-heavy services. It may also push incumbent vendors to accelerate their own agent offerings, compressing timelines for widespread deployment of semi-autonomous systems inside enterprises.

The flip side is capital rotation. If investors conclude that traditional SaaS and IT services face structural margin pressure from AI, they will look more aggressively at companies that own compute, frontier models, or agent orchestration layers. That could further concentrate resources in a small set of labs and infrastructure providers – speeding technical progress even as it makes the ecosystem more fragile.