Google has introduced a specialized bug bounty program aimed at detecting vulnerabilities in its artificial intelligence (AI) products, further enhancing its existing Vulnerability Reward Program. This initiative allows security researchers to earn rewards of up to $30,000 for identifying critical flaws in products like Search, Gemini apps, and Workspace. The program seeks to address “rogue actions” in AI, where systems could behave unexpectedly, such as leaking private data or executing incorrect commands. Higher rewards, up to $20,000, are available for vulnerabilities in key products, with exceptional reports potentially earning an additional $10,000. Google has emphasized that issues related to generated content, like hate speech or copyright violations, should be reported through internal feedback tools for targeted improvements. Researchers have already garnered over $430,000 in AI-related risk exposure, reflecting the ongoing commitment to enhancing AI security by identifying and mitigating potential threats effectively.
Source link
Share
Read more