摘要
目前,以ChatGPT为代表的生成式大模型技术引领着工业人工智能技术的发展趋势,推动着新型工业化的不断深化和提升。但大模型本质上是基于深度学习的一个大型模型,也面临着人工智能安全方面的诸多威胁。本文首先介绍了大模型的发展历程及核心技术、国内外工业大模型研究现状,然后引出了工业大模型的模型安全问题,接着介绍了工业大模型的4种典型应用场景,最后给出了解决工业大模型安全问题的建议。
Currently,the generative large model technology represented by ChatGPT is leading the trend of industrial artificial intelligence technology and driving the continuous deepening and upgrading of the new industrialization.However,the large model is essentially a large-scale model based on deep learning,and it also faces many threats to artificial intelligence security.This paper first introduces the development process and core technologies of the large model,and the current research status at home and abroad of large industrial model,and then leads to the model security problems of industrial large models.After that,four typical application scenarios of large industrial model are introduced.Finally,some suggestions are given to solve the security issues of large industrial model.
作者
芦存博
左璇
金博
张恒硕
张乃柏
Lu Cunbo;Zuo Xuan;Jin Bo;Zhang Hengshuo;Zhang Naibai(School of Integrated Circuits,Beijing University of Posts and Telecommunications,Beijing 100088,China;China Industrial Control Systems Cyber Emergency Response Team,Beijing 100040,China;China Electronics Technology Group Corporation,Beijing 100036,China;Special Police Academy of the PAPF,Beijing 102211,China)
出处
《新型工业化》
2024年第7期85-95,共11页
The Journal of New Industrialization
关键词
工业大模型
新型工业化
人工智能安全
Large Industrial Model
New Industrialization
Artificial Intelligence Security