Meta Says Foundation Model Gives Virtual Embodied Agents Human-Like Movements

Meta has unveiled a foundation model for controlling the behavior of virtual embodied humanoid agents, saying it will enhance the Metaverse.

The new Meta Motivo enables these agents to learn human-like movements and perform complex tasks, the company said in a Thursday (Dec. 12) blog post.

Meta Motivo also adjusts to gravity, wind and other changes in the environment, without being trained for them, according to the post.

“In the future, we believe this research could pave the way for fully embodied agents in the Metaverse, leading to more lifelike NPCs, democratization of character animation, and new types of immersive experiences,” the post said.

Meta said in April that its capital expenditures on artificial intelligence and the Metaverse-development unit Reality Labs will range between $35 billion and $40 billion by the end of the year.

Those figures were $5 billion more than the company initially forecasted for developing new AI products for consumers, developers, businesses and hardware manufacturers, PYMNTS reported at the time.

“We’re building a number of different AI services, from our AI assistant to augmented reality apps and glasses, to APIs [application programming interfaces] that help creators engage their communities and that fans can interact with, to business APIs that we think every business eventually on our platform will use,” Meta CEO Mark Zuckerberg said April 24 during the company’s quarterly earnings call.

Meta also said Thursday that it launched a comprehensive framework for neural video watermarking called Meta Video Seal, which is designed to verify video origins at a time when AI-generated content is on the rise.

This tool can add a watermark and an optional hidden message into videos in a way that is imperceptible to the naked eye, resists common video editing efforts and can later be uncovered to determine the origin of the video, according to the blog post.

“While AI tools can help bring the world closer together, it’s important that we implement safeguards to mitigate the risks of imitation, manipulation and other forms of misuse that can undermine their benefits,” the post said. “Post-hoc watermarking is a crucial step towards better traceability for content and AI models.”

The post Meta Says Foundation Model Gives Virtual Embodied Agents Human-Like Movements appeared first on PYMNTS.com.