Journal of Information Security Reserach ›› 2023, Vol. 9 ›› Issue (6): 518-.

Previous Articles     Next Articles

Research on Artificial Intelligence Data Falsification Risk  Based on GPT Model


  • Online:2023-06-04 Published:2023-06-03



  1. (北方健康医疗大数据科技有限公司济南250117)
  • 通讯作者: 孙雷亮 高级工程师.主要研究方向为攻防对抗、安全运营、安全架构设计.
  • 作者简介:孙雷亮 高级工程师.主要研究方向为攻防对抗、安全运营、安全架构设计.

Abstract: The rapid development and application of artificial intelligence technology have led to the emergence of AIGC (Artificial Intelligence Generated Context), which has significantly enhanced productivity. ChatGPT, a product that utilizes AIGC, has gained popularity worldwide due to its diverse application scenarios and has spurred rapid commercialization development. This paper takes the artificial intelligence data forgery risk as the research goal, takes the GPT model as the research object, and focuses on the possible causes of data forgery and the realization process by analyzing the security risks that have been exposed or appeared. Based on the offensive and defensive countermeasures of traditional cyberspace security and data security, the paper makes a practical study of data forgery based on model finetuning and speculates some data forgery utilization scenarios after the widespread commercialization of artificial intelligence. Finally, the paper puts forward some suggestions on how to deal with the risk of data forgery and provides directions for avoiding the risk of data forgery before the largescale application of artificial intelligence in the future.

Key words: artificial intelligence, artificial intelligence generated context, GPT, ChatGPT, data falsification, offensive and defensive confrontation

摘要: 随着人工智能技术的快速发展应用,人工智能生成内容(artificial intelligence generated context, AIGC)的出现极大地解放了生产力,以ChatGPT为代表的产品风靡全球,其多样化的应用场景催动商业化迅猛发展.以人工智能数据伪造风险为研究目标,将GPT模型作为研究对象,通过分析其已经暴露或出现的安全隐患,重点研究数据伪造可能出现的原因及其实现过程.结合传统网络空间安全、数据安全攻防对抗方法,对基于模型微调导致数据伪造的实践进行了研究,推测人工智能广泛商业化后部分数据伪造利用场景.最后提出应对数据伪造风险的方法和建议,为将来人工智能大规模应用前规避数据伪造风险提供参考.

关键词: 人工智能, 人工智能生成内容, 生成式预训练模型, ChatGPT, 数据伪造, 攻防对抗