Social media giant Meta is looking to bring “entirely new types of gameplay” to its metaverse by forging generative artificial intelligence tech with virtual, augmented and mixed-reality games. 

According to a recent job listing, Meta wants to research and prototype new “consumer experiences” using generative AI, such as games that are “non-deterministic, personalized, and change every time you play them.”

“This is a nascent area but has the potential to create new experiences that are not even possible to exist today.”

According to Meta, the successful candidate will initially work in its metaverse-building Reality Labs division but will also work with company leadership. The pay would be around $347,000 a year plus a bonus, equity and benefits.

Meta's job description for the new generative AI role. Source: Meta

Meta notes the role’s focus will initially be on Horizon, its ecosystem of mixed-reality games built on its integrated game creation system, but it could expand to include non-Meta platforms such as mobile and PC. 

Another key aspect of the role would be to create tools that could accelerate content creation on Meta’s platforms.

“The Generative AI landscape is evolving quickly, especially for content creation. Entirely new paradigms are being invented, and Meta is investing significantly into leading the industry,” Meta wrote.

Bobber Bay Fishing game on Horizon Worlds. Source: Meta 

It comes as Meta has sunk billions in its metaverse division over the years, which saw a $3.8 billion loss in the first quarter this year. 

During an earnings call at the time, Meta CEO Mark Zuckerberg said there would be an “increasing amount” of its Reality Labs work that will go toward serving its AI efforts.

Brazil regulator bans Meta from mining data to train AI models

Meanwhile, Meta’s AI efforts have suffered a blow in Brazil after the national data protection watchdog determined Meta is not allowed to use Brazilians’ personal data to train its AI models.

Brazil’s National Data Protection Authority ordered the “immediate suspension” of Meta’s new privacy policy, which included the use of photos, audio, videos and posts to train AI models.

It said the preventative measure would “avoid serious and irreparable damage” to users, and noncompliance would come with a $8,800 (50,000 Brazilian real) fine per day.

In a statement to local outlet Núcleo, Meta said it was “disappointed” by the decision and that the move would delay the arrival of AI benefits to the Brazilian people.

Brazil is a large market for Meta’s social media platform Facebook, with 102 million active users, according to the regulator.