Meta, the social media giant, has shifted its focus to artificial intelligence (AI) as its primary driver for development after its metaverse ambitions fell short.
The company’s latest project, the Segment Anything Model, was recently announced, allowing users to identify specific items in an image with just a few clicks. Although still in demo mode, Meta claims that the Segment Anything AI system can already isolate one or more items from an image by identifying the pixels that make up each object.
According to Meta, creating an accurate segmentation model for specific tasks requires specialized work by technical experts with access to AI training infrastructure and large volumes of carefully annotated data.
To this end, the company trained the Segment Anything AI system on over 11 million images, leading to impressive capabilities to handle different types of images that were not seen during training.
While generative AI is an AI system that generates text, images, or other media in response to prompts, Segment Anything is still in its research phase, with no plans to use it in production. Nonetheless, there are concerns related to privacy in the potential uses of artificial intelligence, particularly regarding facial recognition.
Meta’s shift to AI comes after the company abandoned its plans to launch a metaverse in February and laid off over 10,000 workers after ending its Instagram NFT project. The company is now making Segment Anything available for the research community under a permissive open license, Apache 2.0, that can be accessed through the Segment Anything Github.
