Embracing AI at Workplace: Navigating the Challenges and Practical Implications for Employers

Artificial Intelligence (AI) is rapidly transforming industries worldwide, bringing significant changes to the workplace and raising important legal considerations for employers. This article explores the latest AI legislative developments in China, the associated employment law challenges, and the practical implications for employers.

Overview of AI governance in China

1. Legislative developments
   China has been proactive in establishing a regulatory framework for AI. In June 2023, the State Council of China announced a legislative plan to draft a comprehensive Artificial Intelligence Law, and in May 2024, the draft Artificial Intelligence Law was also included in the State Council's legislative plan for 2024.
   
   Apart from the legislative plans, China also has made significant progress in AI governance, introducing a series of legal frameworks. In December 2021, the Cyberspace Administration of China (CAC), the Ministry of Industry and Information Technology (MIIT), the Ministry of Public Security (MPS), and the State Administration of Market Supervision and Administration (MSA) jointly promulgated the Provisions on the Administration of Algorithm-generated Recommendations for Internet Information Services (Algorithm Provisions). In November 2022, the CAC, the MIIT, and the MPS jointly issued the Provisions on the Administration of Deep Synthesis of Internet-based Information Services and in July 2023, the CAC together with six other authorities jointly promulgated the Interim Administrative Measures for Generative Artificial Intelligence Services (Interim Measures). These regulations address various issues such as data privacy, algorithm transparency, and ethical AI use, aiming to create a robust framework for the development and deployment of AI technologies, ensuring they are used ethically and responsibly. In particular, Article 20 of the Algorithm Provisions explicitly emphasized that an algorithm-recommended service provider which provides workers with job scheduling services shall protect workers' lawful rights and interests in obtaining labour remuneration, and improve algorithms on platform order distribution, composition and payment of remuneration, working hours, rewards and punishments, etc.
   
   
2. Government action on work scheduling AI
   On 24 November 2024, the CAC, together with the MIIT, the MPS and the MSA, jointly launched a campaign to enhance governance over algorithm issues of online platforms, running until 14 February 2025. The campaign is divided into three stages. In the first stage (from 24 November 2024, to 31 December 2024), companies are required to self-examine and self-correct on key issues, including exploiting gig workers’ rights by leveraging algorithms to force couriers to work faster and more efficiently, which has been a hot topic in China with this year's summer box-office hit (Upstream) highlighting the struggles of gig workers. In the second stage (from 1 January 2025, to 31 January 2025), the companies' self-examination results will be verified by authorities. By 14 February 2025, the CAC will work with other departments to complete an in-depth assessment of the effectiveness of the campaign. Though the campaign is more targeted to big tech giants, this also showcases the government's eagerness to create a more equitable labour market with protections for all workers, regardless of employment status.

Challenges and practical implications for employers using AI in the workplace

Though there have not been separate employment regulations regarding AI use in the workplace, nor many widely publicised court cases specifically involving AI in employment disputes, there are several notable challenges and some practical implications in response to the ever-changing AI governance landscape in China.

1. Regulatory compliance requirements
   Navigating the evolving regulatory landscape is a significant challenge for employers. Employers must stay updated on new AI laws and ensure their AI systems comply with the latest regulations to avoid legal issues, including conducting security assessments and registering algorithms where required.
   
   
2. Data privacy and security
   AI-powered tools can enhance workplace productivity but also raise privacy concerns. Employers must balance the benefits of improving productivity with the need to respect employee privacy rights. With AI systems processing vast amounts of employee data, ensuring compliance with data protection laws including but not limited to the Personal Information Protection Law (PIPL) is crucial. Employers must implement stringent data security measures and specific required actions include:
   • Employee Consent: Employers must obtain separate consent from employees if (a) their data is to be transferred to third parties, (b) sensitive personal information is processed, or (c) the AI service provider is based outside of China.
   • Personal Information Protection Impact Assessment: If the using of AI entails provision of employee data to a third party or cross border transfer of employee data, employers shall conduct a personal information protection impact assessment.
   • Service Agreement. If an employer uses an AI service vendor and provides employees data to the vendor, it shall reach agreement with the third party clearly stating the rights and obligations of both parties regarding the protection of the employees' personal information in a written agreement such as a service agreement with the vendor.
   • Cross-border Requirements. Depending on the data processing details, the employer may need to comply with additional cross-border data transfer requirements, which may involve regulator-led security assessments, standard contractual clauses, or certification. Please refer to our previous alert for details.
     
     
   Given the PIPL sets out a legal ground to process data based on human resource management and contract conclusion, it is also advised to include using AI tools into such exemption, which shall be clearly defined in employment contracts to enable the employer to process employee data based on this ground.
   
   
3. Trade secret
   With the development of AI technology, employees might accidentally input the company's trade secrets into AI models like ChatGPT, thereby leaking the company's trade secrets. Therefore, companies need to be vigilant about the risks posed by generative AI and should advise employees to avoid inputting sensitive information into external AI models.
   
   
4. Job displacement and skill development
   It is inevitable that the integration of AI in the workplace will lead to elimination of certain job positions, particularly in sectors like manufacturing and customer service. Currently, with limited cases, the judiciary institutions have not yet formed a uniform view regarding the replacement of workers by employers through the introduction of AI. In most cases, employers claimed that the dismissal of employees due to adoption of AI constitutes a "significant change in objective circumstances", thus shall be a legitimate termination cause. Although the basic facts of these cases were similar, the rulings of courts in the different cities varied. Except that Shanghai court supported the employer's claim, all the other courts ruled against the employers. Given courts may vary in their interpretation of whether AI-induced changes meet the criteria for significant changes that justify termination, there would be growing uncertainty about the legal grounds for dismissing employees.
   
   Meanwhile, as AI transforms job roles, employers need to invest in reskilling and upskilling their workforce, which helps employees adapt to new technologies and ensures that the company remains competitive. Employers are advised to invest in training programs to equip their workforce with the necessary skills to work with AI systems. Engaging employees in discussions about AI implementation can help alleviate concerns and foster a collaborative work environment.
   
   
5. Bias and fairness
   AI systems may inadvertently perpetuate biases, leading to discriminatory practices in hiring, promotions, and other employment decisions. Article 4 of the Interim Measures explicitly prohibits discrimination during the provision and uses of generative AI services and encourages effective measures to be taken to prevent discrimination based on ethnicity, religious belief, nationality, region, sex, age, occupation, or health. Though the Interim Measures only set out the liabilities for generative AI service providers without specifying penalties for using AI unproperly such as using discriminatory AI, the non-compliance in using AI may lead to violations of other laws and regulations. For instance, if the AI used by an employer for recruitment is biased against female applicants on the basis of marital status or the number of children they have, it would be a form of employment discrimination prohibited under the Employment Promotion Law and the Circular on Further Regulating Recruitment Practices to Promote Female Employment, which is legally prohibited. Employers must ensure that AI tools are designed and used in a way that promotes fairness and equality. Employers should adopt ethical AI practices, ensuring transparency and fairness in AI decision-making processes to mitigate any biases. Employers who fail to do so may encounter adverse consequences, such as reputational damage, legal disputes, and potential regulatory actions.

Conclusion

China's advancements in AI are reshaping the workplace, bringing both opportunities and challenges for employers. Navigating the complex legislative and regulatory landscape is crucial for leveraging AI's benefits while ensuring compliance and protecting employee rights. By monitoring and adopting the latest ethical and responsible practices, employers can successfully integrate AI into their operations and drive sustainable growth.