Meta to watermark its AI-generated content
Meta said that it will add an invisible watermarking to the AI-generated content produced by its AI chatbot. The feature is a part of the company’s responsible development of AI tools, and safety and transparency efforts. The Facebook and Instagram parent company said that in the coming weeks, it will add invisible watermarking to the Imagine with Meta AI experience for increased transparency and traceability.
Similar to many other AI chatbots, Meta AI produces images and content in response to user prompts. However, Meta is committed to ensuring that malicious individuals do not perceive the service merely as another tool for deceiving the public.
The latest watermark feature would make it more difficult for a creator to remove the watermark.
Meta has announced its intention to employ a deep-learning model for applying invisible watermarks to images generated through its AI tool. These watermarks, named "Imagine with Meta AI," are invisible to the human eye but can be detected using a corresponding model.
In contrast to traditional watermarks, Meta asserts that its AI watermarks are robust against common image manipulations such as cropping, color adjustments (brightness, contrast, etc.), and screenshots. Although the initial rollout of watermarking services will be for images created with Meta AI, the company plans to extend this feature to other Meta services that incorporate AI-generated images.
See What’s Next in Tech With the Fast Forward Newsletter
Tweets From @varindiamag
Nothing to see here - yet
When they Tweet, their Tweets will show up here.