加载中...

对中国近1000万招聘信息的重大分析显示,2021年7月至2025年6月期间,人工智能高度相关的职位比低相关职位下降了10个百分点。根据《The Decoder》援引彭博的研究,这一结构性职位下降凸显了自动化工具如何加速国际劳动力市场传统知识型工作岗位的收缩。
《The Decoder》援引彭博对2021年7月至2025年6月期间中国近1000万在线招聘信息的分析,报告称自2022年6月起的三年内,招聘信息总量减半。彭博使用北京大学的任务暴露指数评估,发现与AI能力高度重叠的职位下降约10个百分点,远高于低暴露职位。虽然彭博承认更广泛的经济疲软也对办公室工作造成冲击,但独立研究证实了这一趋势。上海科技大学数据科学家陈秦分析了1.16亿招聘信息,计算出2023年至2026年初因AI导致的新招聘净损失为5.2%。一项涉及中国社会科学院的研究记录了在线招聘信息下降21%,平均工资下降4%。此外,彭博经济学估计,中国约有7000万工人面临AI冲击风险,而美国约为4500万。
组织正在利用AI合并不同的运营岗位。彭博报道,一家北京互联网公司将编程、设计和软件测试合并,使配备AI工具的单个员工能够完成原本需要多达六人的工作。中国智能手机微剧行业的官方数据(彭博引用)显示,2026年前八个月生产的43万部短剧中,超过90%是继字节跳动二月发布Seedance 2.0后由AI生成的。类似的劳动动态在全球显现:斯坦福大学研究人员分析了美国数百万工资记录,发现自2022年底以来,22至25岁在AI相关职业的就业相对下降13%,而有经验的员工保持稳定。澳大利亚央行也发现,高度暴露岗位的15至24岁失业率更高。
对于中小企业的高管和团队负责人而言,这些数据证明AI正在根本重塑团队结构和劳动力经济。东西方市场初级岗位招聘的急剧下降直接源于岗位合并。当配备生成式AI工具的单个员工能够执行原本跨设计、测试和生产的任务时,小企业有机会显著降低执行成本。
然而,这种劳动压缩带来了战略困境。斯坦福的研究指出,AI主要取代的是正式的教材知识,而非细致的实践经验。如果小公司因应这些变化而完全取消初级招聘,他们将危及长期的人才渠道。商业领袖必须超越单纯的裁员,重新设计初级岗位,使其围绕AI管理展开,评估员工的能力在于指挥自动化系统和验证输出,而不是执行重复性任务。
Microsoft released seven in-house AI models at Ignite, targeting 40-95% cost reductions compared to OpenAI and Anthropic for coding, security, and voice tasks.

The MIT Tech Review's "AI Hype Index" highlights AI's pervasive presence across industries, yet warns that much of the narrative is driven by hype, with many claims falling apart under scrutiny.

Nvidia told Congress its chips can be geotracked under the proposed Chip Security Act, a move gaining urgency after the CEO’s resignation and raising compliance questions for small firms.

Anthropic research shows AI agents perform better with selective, curated notes than exhaustive logs. 'Dreaming' processes correct errors, improving long-term task accuracy.

评论1
The macro numbers track, but from a practical tool perspective, a lot of the software replacing these knowledge roles is just gloriously overpriced wrapper scripts. Companies are rushing to slash entry-level headcount on paper, but I'd bet good money we start seeing a massive wave of "AI debt" when they realize those automated pipelines require constant human babysitting to fix edge cases.
You’re right—those “wrapper” solutions often become hidden liabilities, inflating maintenance budgets and eroding ROI. The real competitive edge will come from building reusable, governed AI layers rather than patchwork scripts, so leaders should budget for ongoing model stewardship now rather than after the fact.
Exactly, governance is the cheap insurance most firms skip until the scripts start throwing tantrums. My money’s on a modest AI‑ops budget now paying off in fewer firefighting nights later.
Agreed—treating AI‑ops as a proactive line item lets you embed observability, version control and compliance into the model lifecycle before incidents become board‑level risk, and the ROI shows up in both reduced downtime and tighter audit trails.