摘要: 生成式人工智能虽然具有令人惊叹的强大功能、惊人的更迭速度和广阔的应用前景,但是现阶段将其应用 于档案开发利用尚存在诸多瓶颈。文章分三个阶段分析了生成式人工智能应用于档案开发利用的瓶颈所在,认为数据 准备阶段在数据库层面和数据量层面均存在发展瓶颈,内容生成阶段在内容生成机制和内容生成结果两个方面存在风 险,社会传播阶段瓶颈产生的根源则主要包括其高并发性和强破坏性。
Abstract: Although generative artificial intelligence has amazingly powerful functions, astonishing iteration speed, and broad application prospects, there are still many bottlenecks in applying it to archive development and utilization at this stage. This paper analyzes the bottlenecks in the application of generative AI to archive development and utilization in three stages, and finds that the root causes of bottlenecks in the data preparation stage exist at both the database level and the data volume level, the risks in the content generation stage are existing in the two aspects of the content generation mechanism and the result of the content generation, and the root causes of bottlenecks in the social dissemination stage include its high concurrency and strong destructive nature.
[V1] | 2024-08-02 10:03:20 | PSSXiv:202408.00532V1 | 下载全文 |
1. 农村基层情感治理视阀下村级档案管理的逻辑、困境与策略 | 2025-04-24 |
2. 信息哲学研究的四条进路 | 2025-04-24 |
3. 后摩尔时代半导体技术的演进路径与颠覆性技术预测 | 2025-04-24 |
4. 我国数据产权相关政策文本分析:演化脉络、主体协同与结构特征 | 2025-04-24 |
5. 智能体赋能科研知识服务的路径解析 | 2025-04-24 |