The hottest buzzword in enterprise tech is "AI agent." The inconvenient truth: most of what's being sold under that label barely qualifies.

Gartner predicts that more than 40% of agentic AI projects — tools built to reason, plan, and take action without human hand-holding — will be scrapped or significantly reworked by 2027. Of the thousands of companies now marketing AI agent products, roughly 130 are building something that actually meets the bar.

The rest is what the industry is starting to call "agent washing." Vendors are rebranding chatbots and basic automation tools as agents, banking on the fact that most buyers can't tell the difference. Right now, many can't.

The disconnect goes deeper than labeling. McKinsey's 2025 Global Survey on AI found that just 23% of companies have managed to scale agentic AI beyond the pilot stage — meaning roughly three out of four organizations experimenting with agentic AI are stuck testing, spending money without much to show for it.

"Most deployments today use AI in a shallow way — as an assistant that sits alongside existing workflows and processes — rather than as a deeply integrated, engaged, and powerful agent of transformation," Mistral AI CEO Arthur Mensch wrote in a McKinsey report on agentic AI.

The failures aren't random. Thomson Reuters outlined in a fictional "premortem" scenario how agentic AI pilots at law firms could collapse because firms underestimate the sheer difficulty of getting agents to work across different systems, handle ambiguity, and deal with humans who don't communicate in tidy, structured prompts. "Agents are not simply magical plug-n-play pieces," Mensch added. "They must work across systems, reason through ambiguity, and interact with people — not just as tools, but as collaborators."

The spending enthusiasm is cooling accordingly. As the Wall Street Journal reported, "The golden age of unbridled spending on AI software might be behind us, as vendors say it's a lot harder to make a sale than it used to be."

None of this means agentic AI is vaporware. The companies taking it seriously are digging in for the long haul:

  • Expedia warned in its latest 10-K filing that "the rapid emergence and adoption of generative and agentic AI is likely to further intensify competition," flagging the threat of new entrants who might "deploy AI-driven travel search, planning, and booking capabilities more effectively or rapidly than we can."
  • OpenAI is launching 'Frontier Alliances' with consulting giants Accenture, BCG, Capgemini, and McKinsey to push enterprise AI past the pilot phase — a tacit admission that even its best models don't sell themselves.
  • Inside Salesforce, a company survey found employees are producing more with AI tools, but the work itself isn't getting easier. "The gaps suggest people believe AI is enabling them to do more work, but it's not making their work easier," Jason Schloetzer, an associate professor at Georgetown University's business school, told Business Insider.

Valley View

The coming shakeout will thin the herd, and that's a good thing for enterprises tired of wading through snake oil. The real agents — ones that can navigate messy systems and make judgment calls without a human babysitter — will earn their keep. Everyone else will quietly rebrand as "AI-powered workflows" and hope nobody remembers the pitch deck.