Core Viewpoint - The article discusses the challenges and implications of using trade secret law to protect AI innovations, highlighting the tension between the need for innovation protection and the demand for transparency in AI technology [1][5][14]. Group 1: Trade Secret Law and AI Protection - The U.S. Patent and Trademark Office and Copyright Office have issued guidelines limiting patent and copyright protections for AI-generated innovations, making trade secret law a crucial tool for protecting AI-related innovations [1]. - Trade secret law offers broad protection for sensitive information related to AI, such as training data and model weights, which are not covered by patents or copyrights [1][5]. - The application of trade secret law in AI faces challenges, including insufficient technical transparency and potential misuse of legal protections [5][6]. Group 2: Ethical and Social Implications - The rapid development of AI creates a conflict between the insatiable demand for data and the need for transparency, potentially leading to ethical and social issues [2]. - The use of confidential data to train generative AI (GenAI) has been deemed illegal in multiple jurisdictions, posing irreversible leakage risks [3][4]. - High-profile cases, such as data leaks involving major companies, underscore the risks associated with using public AI tools with sensitive data [4]. Group 3: Transparency and Governance Challenges - The "black box" nature of AI technology conflicts with the need for transparency, particularly in high-stakes areas like healthcare and justice, which could lead to severe consequences [6][9]. - The expansion of trade secret protections can lead to monopolistic practices, limiting competition and innovation, particularly disadvantaging small and medium enterprises [7][13]. - A lack of transparency in AI systems can erode public trust and obscure the risks of technology misuse [7][9]. Group 4: Recommendations for Balancing Interests - To mitigate the negative effects of trade secret law on AI innovation, it is essential to redefine its boundaries, ensuring it only covers genuinely valuable and non-public information [13][14]. - Implementing mandatory transparency requirements and establishing a tiered disclosure system for high-risk AI applications can help balance corporate interests with public needs [10][11]. - The introduction of a compulsory licensing system in emergencies can ensure that critical AI technologies serve public interests while providing fair compensation to developers [10][14].

Group 1 - The application of trade secret law in the AI field needs to redefine boundaries to alleviate the tension between technological innovation and public interest, with a fundamental premise being the prohibition of using confidential information to train GenAI and other public AI tools [1][11] - The U.S. Patent and Trademark Office and the U.S. Copyright Office have issued guidelines that restrict AI-assisted innovation's patent and copyright protections, emphasizing the need for significant human contribution for patent eligibility and excluding AI-generated materials from copyright protection [1][3] - Trade secret law provides strong protection for AI-related innovations, covering sensitive information that patents and copyrights cannot, such as data used for training AI systems or model weights, making it a vital tool for companies [1][5] Group 2 - The rapid development of AI technology has created a conflict between the insatiable demand for data, algorithms, and computing power and the need for transparency in AI technology, leading to potential societal and ethical impacts [2][7] - Using confidential data to feed GenAI has been deemed illegal in multiple countries, with significant risks of irreversible data leaks, as evidenced by international cases of data breaches involving major companies [3][4] - The systemic flaws in GenAI technology highlight the need for strict regulations, including prohibiting access to core business secrets by public AI and ensuring non-sensitive data undergoes dual processing for identity masking and business context separation [4][6] Group 3 - The expansive application of trade secret law in the AI sector has led to overprotection and potential abuse, resulting in legal, ethical, economic, and social crises that pose systemic challenges [5][6] - The lack of transparency in trade secret law creates a fundamental conflict with the need for accountability in AI technology, particularly in high-risk areas like healthcare and justice, which could lead to severe consequences [6][8] - The concentration of AI technology benefits among a few major players due to trade secret protections raises market barriers and diminishes competition, adversely affecting innovation and resource distribution [7][9] Group 4 - To mitigate the adverse effects of trade secret law in AI, a re-examination of its applicable boundaries is necessary, focusing on balancing corporate rights with public interests through transparency requirements and limited protection scopes [8][11] - The introduction of a compulsory licensing system is essential to prioritize public interest in emergencies, allowing governments to utilize protected technologies while ensuring fair compensation for providers [9][12] - Promoting collaboration between legal frameworks and technology development is crucial for optimizing trade secret law's application in AI, including the integration of transparency requirements into industry standards [9][12]