ChatGPT Plus Unveils New Image and Voice Features, Raising Privacy Concerns

OpenAI launched new image and voice features for ChatGPT Plus subscribers, allowing the bot to generate responses based on uploaded images and engage in back-and-forth conversation.

The image feature lets users upload a photo and ask ChatGPT questions about it. It can provide recipe suggestions based on ingredient photos, convert handwritten notes to text, identify objects, and more.

ChatGPT was able to write code based on instructions depicted in a whiteboard photo, demonstrating the feature's potential uses.

The new capabilities raise some privacy concerns, like uploading sensitive personal photos. Users are advised to avoid identifying people.

ChatGPT has guardrails limiting its ability to identify humans in photos, though it was able to identify Frida Kahlo by citing a label on the artwork.