Last week, Gartner released a press release ahead of one of their expos, citing survey results that spell out the tough reality of deploying agents in the enterprise:
- “Agent washing” — where vendors rebrand glorified chatbots as agents — is pervasive.
- Mapping workflows and building DIY agents with custom connections in a complex enterprise stack is expensive and time-consuming.
- Combined, these issues lead to poor outcomes and disappointing ROI.
“Most agentic AI projects right now are early stage experiments or proof of concepts that are mostly driven by hype and are often misapplied,” said Anushree Verma, Senior Director Analyst, Gartner. As a consequence, companies will scrap at least 4 out of 10 of these projects.
This aligns with what I’m hearing across my network. The tech is exciting — everyone wants a taste. And there are bright spots. Some agents really do run in the background, doing work while you sleep (like at Steady).
But let’s be clear: a chat/RAG/search tool, while useful, is a feature. It’s not an outcome. It’s not work done. It requires the user to sit at a keyboard and ask for help.
“To get real value from agentic AI, organizations must focus on enterprise productivity, rather than just individual task augmentation,” said Verma.
That’s exactly what we’re focused on at Steady: a shared brain for teamwork. Our agents assemble and deliver context and actionable coordination in the background so you and your team can stay in flow.
The sooner we clear out the hype-driven not-quite-agents, the sooner real impact can begin.
