Google's new policy targets offensive content in AI apps

Google's new AI app policy: Google has come up with new policies requiring applications like ChatGPT and Bing, using generative AI, to implement an in-app system for reporting offensive content. These policies are set to take effect from January 2024, compelling app developers to use user reports to curtail the generation of "restricted content" and discourage "deceptive behaviour." 

This decision comes after incidents such as the use of the AI-powered photo editor app Lensa for generating explicit content in the past. Microsoft's Bing Image Creator, powered by DALL-E 3, also faced criticism when it produced an image of the fictional character Mickey Mouse holding a gun while flying toward two towers.

Furthermore, Google is introducing changes to permissions required by generative AI apps, like ChatGPT. The company stipulates that "apps may only access photos and videos for purposes directly related to app functionality." Consequently, AI apps such as ChatGPT, which typically request photo or video access despite not requiring it for core functionality, will need to adhere to Google's designated system picker.

Another major alteration pertains to how apps display full-screen notifications. Currently, many apps exploit this feature to entice users into making subscriptions or in-app purchases. With the implementation of the new policy, Google plans to restrict this functionality, and apps displaying full-screen notifications will necessitate special app access permission.