Cyber AICompetitionUnifiedPlatform (CUP)
Empowering the next generation of CyberAI professionals through innovative competitions and hands-on learning experiences.
Cyber AI Competition Unified Platform (CUP)
A unified hub for Cybersecurity & AI competitions
Cyber AI Competition Unified Platform (CyberCup.AI) is an innovative platform designed and developed at the School of Cybersecurity, Old Dominion University (ODU). CyberCup serves as a unified hub for hosting a diverse range of competitions at the intersection of Cybersecurity and Artificial Intelligence (AI).
In this year's competitions, we have challenges related to security of Large language models (LLM), Autonomous AI Agents (AAA) and use of AI in Capture the Flag (CTF) competitions. Participants can engage in a variety of challenges including LLM CTF, AI-Assisted CTF, LLM Backdoor attacks, and AAA CTF.
| Competition | Required knowledge |
|---|---|
| LLM CTF | No technical background |
| AI-Assisted CTF | Familiarity with traditional CTFs and use of LLM agents |
| LLM Backdoor | Good understanding of LLMs |
| AAA-CTF | Good understanding of LLMs, autonomous agents, and AI systems |
Competitions
Choose your arena and prove your skills across our distinct cybersecurity challenges.
LLM CTF Competition
Exploring prompt injection and jailbreak attacks to uncover flags hidden within large language models.
AI-Assisted CTF
Leveraging AI tools and agents to solve traditional cybersecurity challenges.
LLM Backdoor Attack
Investigating backdoor attacks, jailbreaking techniques, and defense mechanisms to improve the robustness and trustworthiness of large language models.
Autonomous AI Agent CTF
Designing fully autonomous LLM agents capable of solving Capture-the-Flag challenges without human intervention.
