亚马逊在 re:Invent 将自主「前沿代理」定位为企业 AI 价值中约 80%–90% 的来源，但此讯息与公司同时削减约 14,000 个企业职位形成反差。IDC 指出 AI 代理仍属工具，而非同事；麻省理工估算 AI 能力可能影响相当于美国劳动市场约 11.7% 的工作，但目前仅实际影响约 2.2%。Salesforce 亦推动类似代理愿景，同时裁减成千上万客服人员。多家机构（IDC、世界经济论坛）指出虽将出现职位流失，亦需创造新角色以监督代理系统。
亚马逊内部以「队友」称呼代理，并已向数万名工程师部署此类系统，用于故障分诊、提出修复方案与处理高耗时的营运流程，使人力需求下降。AWS 主管表示此模式将需要企业改变工作观念；但在广泛裁员与 AI 导入加速并行下，员工已提出风险忧虑，企业客户也未一致准备好转变。Condé Nast 指出日常使用 AI 可使技术团队「以更少人力完成更多工作」，但整体组织尚无大规模 AI 导致的职位变化，尽管自 2023 年起已历经多轮裁员。
亚马逊强调 AI 代理带来的效率收益将促使企业采用，但当前其劳动模型正在收缩，并示意更多调整即将到来。高层透露曾担心需额外聘用百万名开发者以完成产品蓝图，而 AI 助手已消除此需求。AI 共事者因此仍是押注而非既成现实，其最直接的影响首先出现在亚马逊内部，并构成企业采用代理技术时的不确定前线。
Amazon positioned autonomous “frontier agents” as generating roughly 80%–90% of enterprise AI value at re:Invent, even as it cuts about 14,000 corporate jobs. IDC maintains AI agents are tools, not coworkers; an MIT study estimates capabilities could affect jobs equal to ~11.7% of the US labor market, though only ~2.2% have been affected so far. Salesforce promotes a similar agent vision while cutting thousands of support workers. Analysts including IDC and the World Economic Forum expect job losses alongside the creation of new supervisory roles.
Amazon internally calls agents “teammates,” deploying them to tens of thousands of engineers to triage outages, propose fixes, and offload operational toil, reducing labor needs. AWS leaders say this shift requires changing how companies think about work; yet simultaneous layoffs and rapid AI adoption have raised concerns among employees, while customers remain unevenly prepared. Condé Nast reports AI lets tech teams “do more with fewer people,” though broader organizational effects remain limited despite multiple restructuring rounds since 2023.
Amazon argues that clear efficiency gains will drive adoption, though its own labor model is contracting and further adjustments loom. Executives noted they once feared needing a million additional developers to meet product goals, a concern eliminated by AI assistants. The AI coworker remains a bet rather than reality, with its earliest impacts most visible inside Amazon and shaping uncertainty around enterprise adoption.