Meta Launches Watermarking Functionality for Select AI Products
On Wednesday, Meta Platforms announced its plans to introduce invisible watermarking to its text-to-image generation product, imagine, along with the Meta AI chatbot in the upcoming weeks. This move aims to improve transparency, according to the parent company of Facebook.
At the end of September, the social media company launched products containing artificial intelligence (AI) for consumers, such as robots that create photorealistic images and smart glasses that answer questions.
“We aim to introduce an invisible watermark to many of our products with AI-generated images in the future,” Meta said in a blog post, adding that it can withstand common image manipulations such as cropping and screenshots.
The success of ChatGPT—a Microsoft-backed OpenAI chatbot—has led companies to harness the power of large language models to create AI-powered products to attract new investors, drive innovation, retain old customers, and engage new ones.
Meta made Meta AI using a custom model based on the powerful Llama 2 large-language model, which the company released for public commercial use in July.
The company is testing more than 20 new ways to use generative AI to improve experiences on its social media platforms, including Instagram and WhatsApp.
The Menlo Park, California-based company is expanding the ability to imagine beyond chat and making it available in the US, while testing to improve search capabilities in several of its products.
