| 關 鍵 詞： |
人工智慧；電子監控；電子網路；員工隱私權 |
| 中文摘要： |
歐盟立法者於 2023 年 12 月達成對草案人工智慧（AI）法的政治協議。歐洲委員會在 2024 年 3 月提出的這份草案 AI 法是全球首要具有約束力的橫向 AI 法規，為歐盟內使用和供應 AI 系統設定一個共同框架。該法案根據”基於風險”的方法，為不同需求和義務量身定制 AI 系統的分類，並提出”不可接受”風險的 AI 系統被禁止。許多”高風險”AI 系統，可能對人們的健康、安全或基本權利造成不利影響，獲得授權，但需要遵守一系列要求和義務才能進入歐盟市場。由於缺乏透明度，導致風險有限的 AI 系統將受到信息和透明度要求的約束，而對人們風險極小的 AI 系統則不需要進一步承擔義務。該法規還為通用人工智慧（GPAI）模型提供具體規定，並為具有”高影響能力”的 GPAI 模型制定更嚴格的要求，這些模型可能對內部市場構成系統性風險並產生重大影響。
臨時協議已獲得歐盟成員國常駐代表委員會和議會的兩個主導委員會的認可。議會將於三月全會期間對最終協議進行全體投票。AI 法案還必須經過理事會的認可並刊登在歐盟官方期刊上才能生效。雇主和其他僱用員工來執行相關工作上之活動，並且讓員工使用許多電子機制，其中有關招聘、員工評估、薪酬、紀律等和保留這些電子機制，例如包括電子追蹤器、監控攝影頭、身體的代謝監測儀、測量裝置和其中的技術，雇主利用這些工具啟用來記錄他們員工的一切活動、聆聽他們的對話、測量績效的時間等方面，並檢測反對組織活動，收集通過的數據等等。因人工智慧演算法方法進入永久保存，電子簡歷也可以識別和預測人的績效以及他們的職業道德、個性、工會傾向、雇主忠誠度和未來醫療保健成本、電子簡歷等等。
由人工智慧生產的各種機制追隨員工從一項工作場所到另一項工作環境，因為他們移動圍繞網路世界無遠弗屆的工作場所，因此人工智慧和電子監控產生一個無形的電子網路，不但入侵員工隱私權，還阻止工會運行，使微妙的雇主形式化加劇就業歧視問題，讓工會變成無法正常運作以及無法賦予勞工相關勞工法之保障。
本文介紹人工智慧在工作場所之運用及其使用如何改變招聘之實踐、評估、補償、控制和解僱員工。然後，專注於人工智慧威脅要破壞員工法律領域之保障：反歧視法、隱私權法、反托拉斯法和勞工法，最後，本文希冀能夠提出建議，為法律上未來提供相關改革和研究之方向。
|
| 英文關鍵詞： |
AI；Electronic surveillance；Internet；Employee's privacy |
| 英文摘要： |
EU lawmakers reached a political agreement on a draft artificial intelligence (AI) law in December 2023. The draft AI law, proposed by the European Commission in April 2024, is the world's first binding horizontal AI regulation that sets a common framework for the use and supply of AI systems within the EU. The bill tailors the classification of AI systems to different needs and obligations based on a "risk-based" approach. Some AI systems that pose "unacceptable" risks are banned. Many "high-risk" AI systems, which can adversely affect people's health, safety, or fundamental rights, are authorized, but need to comply with a series of requirements and obligations in order to enter the EU market. Due to the lack of transparency, AI systems with limited risk will be subject to information and transparency requirements, while AI systems with minimal risk to people will not be subject to further obligations. The regulation also provides specific provisions for artificial general intelligence (GPAI) models and sets stricter requirements for GPAI models with "high impact capabilities" that can pose a systemic risk to the internal market and have a significant impact.
The provisional agreement has been endorsed by the Committee of Permanent Representatives of the EU Member States and the two leading committees of the Parliament. The parliament will vote unanimously on the final agreement during the March plenum. The AI Act must also be endorsed by the Council and published in the Official Journal of the European Union in order to enter into force. Employers and others who hire or engage workers to perform services use a dizzying array of electronic mechanisms to make personnel decisions about hiring, employee evaluation, compensation, discipline, and retention. These electronic mechanisms include electronic trackers, surveillance cameras, metabolism monitors, wearable biological measuring devices, and implantable technology. These tools enable employers to record their workers' every movement, listen to their words, measure the minutes of performance evaluation, and detect oppositional organizing activities.
The data collected is transformed by means of artificial intelligence algorithms into a permanent electronic resume that can identify and predict an individual's performance as well as their professional ethics, personality, union proclivity, employer loyalty, and future health care costs. The electronic resume produced by AI will accompany workers from job to job as they surround the boundless workplace in the cyber space. Thus, AI and electronic monitoring produce an invisible Internet that invade worker privacy as well as deter unionization, enabling subtle forms of employer blackballing, exacerbating employment discrimination. With these demerits, unions become ineffective and obliterate the protections of the labor laws.
This article describes the many ways AI is being used in the workplace and how its use is transforming the practices of hiring, evaluating, compensating, controlling, and dismissing workers. It then focuses on four fields of law in which AI threatens to undermine worker protections：Anti-discrimination law, privacy law, antitrust law and labor law. Finally, this article maps out an agenda for future law reform and research.
|
| 目 次： |
壹、前言
貳、人工智慧與電子監控
一、以演算法成立的人力資源
二、數據演算及深度學習
三、機器人與人工智慧
四、計算機之計算和語音辨識
參、人工智慧在工作場所產生之問題
一、招募聘僱
二、就業歧視
三、偏見產生
四、人工智慧減少偏見
肆、人工智慧在工作場所的法律問題
一、員工就業隱私權保障
二、勞工法問題
伍、結論與建議
|
| 相關法條： |
 |
| 相關判解： |
|
| 相關函釋： |
|
| 相關論著： |
|