nthropic’s enterprise push. The company says tech is its biggest source of business revenue, with financial institutions next, and it has been demoing finance-focused agents to show how automation can fit into regulated workflows. Competition is getting sharper. With backing from Google and Amazon, Anthropic has resources and distribution, but customers are increasingly comparing agent platforms on reliability, security, and measurable time saved – not just model quality.
Why should I care?
For markets: Software winners may be the ones that control the workflow layer.
As agents get better at remembering context and coordinating specialist agents, they start to look less like chatbots and more like a layer that sits on top of – or replaces parts of – workplace software. That raises a hard question for investors: how durable is “seat-based” pricing, where companies pay per user, if more work gets done by automated systems? Even before revenue shifts show up in results, that uncertainty can pressure valuations across software-as-a-service businesses.
The bigger picture: AI is shifting from clever answers to accountable digital work.
“Dreaming” points to a broader trend: companies want systems that improve with use, remember preferences safely, and operate across many steps, not just a model that responds to prompts. But more autonomy also means more scrutiny around governance – the guardrails, logs, and approvals that make it clear what the system did and who is responsible when something goes wrong. The firms that solve that trust-and-control problem fastest are likely to set the pace for adoption.