On November 13, 2025, the first edition of LLM4Sec, the Workshop on the Use of Large Language Models for Cybersecurity, took place in Washington, DC, USA. Organized by Antonino Rullo (ICAR), Giuseppe Manco (ICAR), and Luca Caviglione (IMATI), LLM4Sec aims to create a forum for researchers interested in understanding how LLMs can support defense activities, threat analysis, security process automation, and the detection of advanced attacks. In this first edition, hosted by the ICDM 2025 conference, 12 research contributions were presented, covering a wide range of topics, including the use of autonomous and multi-agent systems, generative and analytical AI for vulnerability discovery, and malware and attack detection. The presentations and discussion sessions highlighted both the significant potential of LLMs in addressing emerging cybersecurity challenges and the critical issues and risks associated with their use, fostering a constructive dialogue between academia and industry. This inaugural edition of LLM4Sec marked an important step toward establishing a dedicated community at the intersection of artificial intelligence and cybersecurity, laying the foundation for future collaborations and scientific advancements in the field.

web site: https://llm4sec-workshop.github.io/