Scholarlink

This research shows the possibility of applying large language models as a method of cyberattack. Large language models are trained on large amounts of data and they can provide answers which can be easily used for malicious purposes. Answers that are provided by language models, attackers are using for malicious purposes so language models become the main helpers in making cyberattacks. The research shows a brief summary of methods that are used for cyberattacks using the help of large language models and it shows the efficiency of implementing attacks.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The Possibility of Applying Large Language Models as a Cyberattack Method

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ivan Cvitić,
  • Dragan Peraković,
  • Anđela Stanišić

摘要

This research shows the possibility of applying large language models as a method of cyberattack. Large language models are trained on large amounts of data and they can provide answers which can be easily used for malicious purposes. Answers that are provided by language models, attackers are using for malicious purposes so language models become the main helpers in making cyberattacks. The research shows a brief summary of methods that are used for cyberattacks using the help of large language models and it shows the efficiency of implementing attacks.