The Hacker Perspective on Generative AI and Cybersecurity

This HackerOne report summarizes ethical hackers' perspectives on security risks introduced by Generative AI and LLMs — highlighting prompt injection, agent access control, AI-assisted attack automation, and the evolving role of hackers and bug bounty programs in identifying and mitigating these vulnerabilities — and recommends testing, stronger access controls, and leveraging hacker communities to secure AI deployments.