這一集從《經濟學人》的一篇文章出發,聊聊一個看起來很矛盾的現象:AI 已經讓一些工作消失,但同一時間,資料中心、電力、營造、AI 相關職缺也正在快速增加。
我們會進一步拆解經濟學裡的 displacement effect 和 reinstatement effect,看看 AI 為什麼可能先取代「工作裡的某些任務」,卻不一定立刻讓整個職業消失,也會談到 Noy & Zhang、Brynjolfsson 等人的研究,看看 AI 到底有沒有真的提高生產力。
但真正值得注意的,無論總工作數最後是增加還是減少,哪些人會最先受到衝擊呢?年輕人會不會更難進入職場?新人原本用來累積經驗的工作,會不會先被 AI 吃掉?即使新工作真的出現,原本的勞工又能不能順利轉過去?
這集想聊的,就是 AI 時代下,就業市場真正正在發生的改變。
如果有任何問題,或想聽什麼內容,歡迎大家聯繫以下的信箱:
[email protected]
也歡迎大家上 YouTube 搜尋《經濟有謂 Econ Savor》,
按下訂閱並跟身邊的親友分享。
參考文獻:
Acemoglu, Daron & Pascual Restrepo (2019). “Automation and New Tasks: How Technology Displaces and Reinstates Labor,” Journal of Economic Perspectives, 33(2), 3–30. Autor, David, Caroline Chin, Anna Salomons & Bryan Seegmiller (2024). “New Frontiers: The Origins and Content of New Work, 1940–2018,” The Quarterly Journal of Economics, 139(3), 1399–1465. Brynjolfsson, Erik, Bharat Chandar & Ruyu Chen (2025, revised 2026). “Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence,” Stanford Digital Economy Lab Working Paper. Brynjolfsson, Erik, Danielle Li & Lindsey R. Raymond (2025). “Generative AI at Work,” The Quarterly Journal of Economics, 140(2), 889–942. Gmyrek, Paweł, Janine Berg, Karol Kamiński, Filip Konopczyński, Agnieszka Ładna, Balint Nafradi, Konrad Rosłaniec & Marek Troszyński (2025), “Generative AI and Jobs: A Refined Global Index of Occupational Exposure,” ILO Working Paper 140. Humlum, Anders & Emilie Vestergaard (2025, revised 2026). “Still Waters, Rapid Currents: Early Labor Market Transformation under Generative AI,” NBER Working Paper No. 33777. Noy, Shakked & Whitney Zhang (2023). “Experimental Evidence on the Productivity Effects of Generative Artificial Intelligence,” Science, 381(6654), 187–192. The Economist (2026), “The jobs apocalypse is postponed. An AI jobs boom is here,” September 12–18, 2026.
--
Hosting provided by SoundOn