IWGB Game Workers' Branch、Game Workers Unite Ireland、西班牙CSVI以及意大利FIOM-CGIL米兰工作委员会等机构近日联合发表了一份宣言,明确反对在游戏开发过程中使用生成式人工智能(GenAI)1。这份文件指出,GenAI的普及正在导致工作条件恶化,使从业者产生“认知投降”现象,进而引发疲劳感与职业疏离1。此外,该技术还被认为加剧了职场剥削,并存在窃取人类创作者劳动成果的风险1。
在就业层面,宣言强调GenAI直接削减了初级职位的需求,阻碍了新入行者进入游戏行业,甚至出现部分公司强迫员工训练最终将替代其自身岗位的AI模型的情况1。针对版权与伦理问题,各方指责大多数GenAI模型基于未经授权的第三方数据进行训练,这不仅引发了严重的西方中心主义偏见,还带来了文化同质化的风险1。为了应对这些挑战,宣言向雇主提出建议,呼吁完全避免使用GenAI,或在代码规范中禁止其在早期开发阶段的应用,同时严禁在未获明确同意和补偿的情况下将员工作品用作训练数据1。对于监管机构,文件要求建立更严格的法律标准,禁止在未获许可且无公平补偿的前提下,将个人作品提交至任何GenAI数据集1。该宣言还引用了Shaw & Nave (2026)关于“认知投降”的研究,以及Cheng et al. (2026)关于谄媚型AI减少亲社会意图的研究作为理论支撑1。
A coalition of trade unions and works councils across Europe, including the IWGB Game Workers' Branch, Game Workers Unite Ireland, Spain’s CSVI, and the FIOM-CGIL Milan Work Council, has issued a joint declaration strongly opposing the use of generative artificial intelligence (GenAI) in game development. 1 The group argues that GenAI implementation leads to deteriorating working conditions, undermines worker ownership, exacerbates workplace exploitation, and results in the theft of creative output. 1 By citing research from Shaw & Nave (2026), the declaration describes a phenomenon termed "cognitive surrender," which causes workers to feel fatigued and alienated. 1
The unions highlight significant negative impacts on employment, noting that GenAI directly reduces the demand for entry-level positions and creates barriers for new professionals entering the industry. 1 In some instances, companies have reportedly forced employees to train models that would eventually replace them. 1 Ethical concerns are also central to the declaration, with critics pointing out that most GenAI models are trained on unauthorized third-party data, posing risks of severe Western-centric bias and cultural homogenization. 1 Additionally, the document references Cheng et al. (2026) regarding sycophantic AI reducing prosocial intent. 1
Specific recommendations were made to various stakeholders. Employers are advised to either completely avoid using GenAI or prohibit its use during early development stages through coding standards, while ensuring employee work is not used as training data without explicit consent and compensation. 1 Regulators are urged to ban the submission of individual works to any GenAI dataset unless permission is granted and fair compensation is provided. 1 The declaration calls on workers, employers, players, and regulators to resist mandatory AI adoption and establish stricter ethical and legal standards to protect human creators. 1
评论
还没有评论,欢迎留下第一条。