文章指出，科技公司以「取代知识工作」作为衡量标准并反复警告白领大屠杀，强化了社会对失业风险的预期；南韩政策幕僚长 Kim Yong-beom 因在 Facebook 发文提出以 AI 利润带来的「超额税收」向全民发放股利而引发争议，后续在市场波动压力下澄清。作者认为此构想或许过早、对晶片产业时点不佳，但若 AI 同时带来高产出与劳动市场冲击，分红成本可能低于民粹反扑的政治代价。
南韩的讨论之所以关键，在于其并非 AI 保守者：Pew 调查显示，仅 16% 的南韩受访者在日常生活中对 AI「更担忧而非更兴奋」，为全球最低；其亦拥有全球最高工业机器人密度，并被 Microsoft 的 AI Economy Institute 视为扩散最成功案例之一。然而，以 Samsung Electronics Co.、SK Hynix Inc. 等为代表的科技成长，并未平均分配收益，形成分配面压力。
相较之下，美国态度转趋负面：Pew 指出，对 AI「更担忧而非更兴奋」者已达约一半，较 2021 年的 37% 上升，并出现抗议资料中心、入门职焦虑乃至暴力事件等回弹。作者主张仅靠再培训不足，AI 降低认知工作的边际成本、削弱工会等谈判工具，会压缩劳动议价；政策上应要求企业揭露裁员的地点、方式与原因，以辨识先被掏空的区域与职涯梯；并可运用 AI 税收增量来降个人税负、或定向支援失业者过渡；同时推动公众财富基金或「AI 股利」（如 Sam Altman、Elon Musk、OpenAI 相关构想所示）以提高民众共享上行的比例，但须警惕通膨与「只发支票」无法满足工作目的感。
The piece argues that when technology makers frame success as replacing human knowledge work and repeatedly warn of a white-collar “bloodbath,” they prime the public to expect mass livelihood losses. In South Korea, presidential policy chief Kim Yong-beom triggered an uproar with a Facebook post suggesting a citizen dividend funded by excess tax revenue from AI profits; after market volatility, he clarified his remarks. The author views the idea as possibly premature and poorly timed for chip champions, yet potentially cheaper than the political cost of a populist backlash if AI delivers both abundance and labor disruption.
South Korea’s debate matters because it is an unusually enthusiastic AI adopter rather than a laggard. A Pew survey reports only 16% of Koreans say they are more concerned than excited about AI’s rise in daily life, the lowest share globally; the country also has the world’s highest density of industrial robots and was highlighted by Microsoft’s AI Economy Institute as a leading diffusion success. Even so, the tech-led growth associated with firms such as Samsung Electronics Co. and SK Hynix Inc. has not distributed gains evenly, keeping inequality and legitimacy concerns salient.
In the United States, attitudes are hardening: about half of Americans are now more concerned than excited about AI, up from 37% in 2021, alongside protests against data centers, entry-level worker anxiety, and even violence. The essay argues “retraining” alone is inadequate because AI can collapse the marginal cost of cognitive work and weaken bargaining tools such as unions, shrinking labor’s leverage as firms aim to generate similar revenue with fewer employees. It calls for data-driven policy—mandatory disclosure of where, how, and why AI-linked layoffs occur—to identify the first regions and career ladders being hollowed out, then target relief; incremental AI-related tax revenue could cut individual taxes or cushion displaced workers. Proposals for public wealth funds or AI dividends (endorsed in various forms by Sam Altman, Elon Musk, and echoed in OpenAI’s recommendations) could broaden participation in the upside, but must be designed beyond “mailing checks” to avoid inflation pressures and preserve a sense of productive purpose.