Adobe has expanded its bug bounty programme to include generative AI, offering rewards to ethical hackers who discover vulnerabilities in its AI platform, Adobe Firefly, and its Content Credentials service. The company believes that as generative AI becomes more integrated into daily life, it is crucial to understand and mitigate the associated risks. By expanding its programme, Adobe hopes to foster an open dialogue and encourage fresh ideas and perspectives in the development of safe and trustworthy AI solutions. The expansion covers issues arising from prompt injection, sensitive information disclosure, and training data poisoning. Content Credentials, an integral part of Adobe’s products, provide transparency regarding the origin of items created using Firefly. Hackers interested in participating can find more information on Adobe’s HackerOne page or apply to join its private bug bounty programme. Adobe will also be present at the Bug Bounty Village at BSides San Francisco and sponsoring a Saturday night party focused on technology’s ethical struggles. Other tech firms, such as Google, have also implemented bug bounty programmes to address the risks associated with generative AI.
Menu
