---
id: "contrarian-rpa-is-bad"
type: "contrarian-insight"
source_timestamps: ["§ Information Systems"]
tags: ["rpa", "automation"]
related: ["claim-screen-clicking-is-flawed"]
challenges: "The massive industry investment in UI-navigating AI agents and advanced Robotic Process Automation."
speakers: ["Harang Ju"]
sources: ["agentic"]
sourceVaultSlug: "hbr-seg-agentic"
originDay: 6
articleStem: "hbr-ext-17-workplace-set-up-for-agents"
sourceUrl: "https://hbr.org/2026/01/is-your-workplace-set-up-for-ai-agents"
sourceTitle: "Is Your Workplace Set Up for AI Agents?"
---
# Screen-clicking AI (RPA) is a dead end.

Much current enterprise excitement around AI involves agents that navigate existing software UIs (like advanced RPA). The author views this entire approach as a fragile, temporary workaround — [[quote-pretending-to-be-human|'pretending to be human']] — rather than a real solution. See the underlying claim [[claim-screen-clicking-is-flawed]].

**Challenges:** the massive industry investment in UI-navigating AI agents and advanced Robotic Process Automation.

**Balanced view (enrichment):** RPA is often the only feasible option in heavily regulated or legacy environments lacking APIs and can deliver robust ROI with governance; many modern platforms blend RPA with APIs and event-driven integration. A domain expert would endorse 'API-first' in principle while treating transitional UI automation as an economically rational bridge technology rather than a strict dead end.
