2026-11-04 · Fonte：Computerworld
Harvard Business Review argues AI agents deserve workforce-style governance — but its study of 1,261 managers shows employee framing makes experienced managers take less responsibility and catch fewer errors.
An addition to the workforce that never sleeps, never asks for a raise and can be cloned before lunch might sound like an ideal hire — except it can confidently make the wrong decision thousands of times before anyone notices. That tension explains why some technology leaders have begun describing AI agents as employees, co-workers or interns rather than software. Once an agent can update records, issue refunds, send communications and act across multiple enterprise systems, deploying it stops looking like installing another application.
Recent Harvard Business Review articles have extended the workplace metaphor into a governance framework. One argues that agents should be managed like a new workforce, complete with defined roles, explicit authority, designated sources of truth, supervision and audit trails. A second goes further, recommending that organizations give agents names and job descriptions, onboard them formally, evaluate their performance, and expand their responsibilities only after they have proven themselves.
The catch is that people may start believing the metaphor. A separate HBR study of 1,261 managers found that when AI was framed as an employee rather than a tool, managers who had hands-on experience working around AI agents took less personal responsibility for outcomes and assigned more responsibility to the AI itself. Those same managers escalated more problems and caught fewer errors — a pattern with direct consequences for oversight and accountability in enterprises scaling agentic deployments.
The framing is already widespread. In the study, 31% of respondents said their companies present AI as a teammate or employee, while 23% reported that agents now appear on organizational or work charts. For IT leaders, the implication is a careful balancing act: adopt the discipline of workforce governance — roles, supervision, auditability — without letting the language of employment erode human ownership of results. Agents can be governed like workers, the argument runs, as long as no one pretends they are human.
Tags: Harvard Business Review · AI Agents · AI Governance · Enterprise IT · Agentic Workforce
萬安算交所 AIXX · AIXX · Exchange verificado · Ir ao mercado