Meta says it will label more AI-generated images on its platform ahead of the 2024 election season. The social media giant is working with tech companies such as Google and OpenAI to design invisible metadata or “watermarks” that identify photorealistic images made with their artificial intelligence tools. Meta, which already labels its own AI-generated images, will roll out the labeling across its Facebook, Instagram and Threads platforms in multiple languages over the coming months. The move comes amid concern that, paired with the distribution power of social media, the images could confuse the public.
- Meta warned it would punish users who don’t disclose when realistic video or audio content was made with AI.
- Meta has joined the dividend club, writes The Wall Street Journal, and while for the company it shows its “confidence in its ability to keep churning out cash,” some investors believe “the move shows Meta’s days of explosive growth are likely in the past.”
- OpenAI will add watermarks to its images generated on the ChatGPT website and the API for the DALL-E 3 model, reports The Verge.