AI 替代岗位 构成违法解雇？｜南财合规周报

Group 1: AI and Copyright Issues - A group of six authors, including John Carreyrou, has filed a lawsuit against six AI companies, alleging unauthorized use of copyrighted books for training large models [2] - In a previous case, Anthropic reached a $1.5 billion settlement with authors over similar copyright infringement claims, with each eligible author receiving approximately $3,000 [2] - The current lawsuit emphasizes that large language model companies should not be able to settle numerous high-value infringement claims at a low cost [2] Group 2: Labor Law and AI Replacement - A case in Beijing ruled that terminating an employee due to their position being replaced by AI constitutes illegal dismissal [3][4] - The employee, Liu, contested the termination after his role in data collection was eliminated in favor of an AI-driven model [3] - The arbitration committee determined that the company attempted to shift the costs and risks of technological advancement onto the employee, which was deemed unlawful [4] Group 3: AI Emotional Support Regulations - The National Internet Information Office proposed new regulations for AI emotional support services, focusing on user interaction and data protection [5] - The draft requires clear identification of non-human interactions and mandates user notifications for prolonged usage [5] - Strict guidelines on training data usage were introduced, prohibiting the use of user interaction data for model training without explicit consent [5] Group 4: Cybersecurity and Platform Governance - OpenAI acknowledged the challenges in defending its AI browser, Atlas, against prompt injection attacks, despite implementing automated defense strategies [7] - Kuaishou experienced a significant security breach, leading to a temporary shutdown of its live streaming feature due to an influx of inappropriate content [9] - The incident raised concerns about the platform's security capabilities, resulting in a more than 3% drop in its stock price [9] Group 5: Content Moderation and Platform Responsibility - The Lijiang Cultural and Tourism Bureau criticized Xiaohongshu for inadequate content moderation, leading to false claims against businesses [8] - Xiaohongshu responded by emphasizing its commitment to authentic user experiences and providing channels for businesses to report disputes [8] - Ongoing investigations have not yet confirmed any violations in user-generated content, but the platform continues to monitor for potential issues [8]