Photo by Gene Samit on Pexels

Google is incentivizing security researchers to find vulnerabilities in its AI systems with a new bug bounty program. The program aims to identify and mitigate risks associated with large language models and generative AI, particularly those leading to malicious or exploitative behavior. Rewards will vary based on the severity and impact of the bug, with the highest payouts reserved for critical vulnerabilities in Google’s core AI-powered products. Lower-tier abuses, such as extracting secret model parameters from systems such as Jules or NotebookLM, will also be rewarded. The top prize of $20,000 is reserved for identifying “rogue actions” on Google’s flagship products Search, Gemini Apps, and core Workspace applications like Gmail and Drive. The reward can potentially increase to $30,000 with multipliers for detailed, high-quality reports.