摘要
在智能电视应用聊天生成预训练转换器(Chat Generative Pre-trained Transformer,ChatGPT)等人工智能(Artificial Intelligence,AI)技术驱动的自然语言处理工具,能够为用户提供更为个性化、便捷化的操作辅助,提升智能电视实际使用体验。然而,ChatGPT还存在算法、模型不透明、应用漏洞频发、数据保护不利以及知识产权侵权等风险。因此,智能电视生产厂商、AI技术提供商应该建立应用、数据防护及知识产权保护等技术体系,避免出现信息安全、数据安全和知识产权安全问题。基于此,阐述ChatGPT应用给智能电视带来的安全风险,探讨应用ChatGPT技术的智能电视需采用的防范措施。
The application of Artificial Intelligence(AI)technology-driven natural language processing tools such as Chat Generative Pre-trained Transformer(ChatGPT)in smart TV can provide users with more personalized and convenient operation assistance,improving the actual use experience of smart TV.However,ChatGPT also carries risks such as opaque algorithms,models,frequent application vulnerabilities,poor data protection,and intellectual property infringement.Therefore,smart TV manufacturers and AI technology providers should establish technical systems such as application,data protection and intellectual property protection to avoid information security,data security and intellectual property security issues.Based on this,this paper describes the security risks brought by the application of ChatGPT to smart TV,and discusses the preventive measures to be adopted for smart TV using ChatGPT technology.
作者
李珣
任艳丽
张震
LI Xun;REN YanLi;ZHANG Zhen(National Industrial Information Security Development Research Center,Beijing 100040,China;Beijing NORTH-STAR Technology Development Co.,Ltd.,Beijing 100089,China;Beijing Electronic Science&Technology Institute,Beijing 102627,China)
出处
《电视技术》
2023年第10期226-228,共3页
Video Engineering
