Pinterest has introduced labels for identifying AI-generated content to its platform. In a blog post shared this week, the company explained that it had been testing the feature over the past few months, and that it is also working on tools to aid users in managing their experience with generative AI. This feature has been rolled out globally, so all users will be able to see the labels.
When browsing Pinterest, the user can now click on an image Pin in close-up to check if it has been created or modified with AI. Images that have been generated using AI will have an “AI modified” label in the bottom left-hand corner to differentiate them from other content.
This feature involves analysing the metadata of an image to determine if it has been generated with AI. Additionally, the company has also developed classifiers to automatically detect AI-generated content, even when there are no obvious markers. Pinterest has acknowledged that this system is not perfect and that there is a possibility for images to be mislabelled. For such cases, the company allows users to contact its support team if they wish to have the label removed from their images.
The AI labels are a first step in dealing with the deluge of AI content on the platform. According to Pinterest, it will soon begin testing the ability to let users filter out Pins made or modified with generative AI. When viewing an AI Pin, the user will find a “see fewer” option in a three-dot overflow menu on the bottom right of the Pin. This option, which may be expanded to other categories in the future, will make Pinterest recommend less AI content, allowing users to have more control over what they want to see on the platform.
(Source: Pinterest blog)
