arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.01435cs.CRcs.AI
收藏

当前通用人工智能模型部署中的根本风险:我们从网络安全中学到了什么(又未学到什么)?

Fundamental Risks in the Current Deployment of General-Purpose AI Models: What Have We (Not) Learnt From Cybersecurity?

CISPA Helmholtz Center for Information Security(CISPA亥姆霍兹信息安全中心)

Mario Fritz

AI总结:

本文审视通用AI模型从语言模型演进为具备工具调用与执行能力的准操作系统所引发的网络安全风险,总结相关评估工作并展望未来机遇与挑战。

摘要:

通用人工智能——例如大语言模型(LLM)——已在广泛的用例中得到快速部署。最令人惊讶的是,它们已从单纯的语言模型发展到聊天机器人,一路达到近乎“操作系统”般的地位,能够控制应用程序的决策和逻辑。工具使用、Microsoft co-pilot/office集成以及OpenAI的Altera,只是自主性、数据访问和执行能力不断增强的少数几个例子。这些方法带来了一系列网络安全挑战。我们重点介绍我们在评估方面所做的一些工作,并概述未来的机遇与挑战。

英文摘要:

General Purpose AI - such as Large Language Models (LLMs) - have seen rapid deployment in a wide range of use cases. Most surprisingly, they have have made their way from plain language models, to chat-bots, all the way to an almost ``operating system''-like status that can control decisions and logic of an application. Tool-use, Microsoft co-pilot/office integration, and OpenAIs Altera are just a few examples of increased autonomy, data access, and execution capabilities. These methods come with a range of cybersecurity challenges. We highlight some of the work we have done in terms of evaluation as well as outline future opportunities and challenges.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑