Meta is a mulling ads and a ‘premium’ version of its AI assistant, Mark Zuckerberg says

One day after Meta rolled out its standalone AI app, Mark Zuckerberg has shared more about how the company plans to eventually monetize its generative AI assistant. During the company’s first quarter earnings call, Zuckerberg said Meta AI could one day show ads and product recommendations. He also hinted at plans for a subscription component for those who want a more “premium” version of the assistant.

“I think that there will be a large opportunity to show product recommendations or ads, as well as a premium service for people who want to unlock more compute for additional functionality or intelligence,” Zuckerberg said.

He added that for now the company is more focused on growing Meta AI’s usage. (He announced yesterday that Meta Ai had reached “almost” 1 billion monthly users.) “I expect that we’re going to be largely focused on scaling and deepening engagement for at least the next year before we’ll really be ready to start building out the business here,” he said.

Zuckerberg’s comments — just one day after Meta introduced its standalone AI app — underscores how important the assistant is to the company. The Facebook founder has repeatedly said he wants Meta AI to be the most used AI assistant in the world, and he said on Wednesday’s call that a standalone app would be particularly important for attracting US users.

Meta’s strategy for monetizing the assistant in many ways mirrors its approach to Threads, which only just began its early experiments with ads this month long after it reached hundreds of millions of users. Speaking of Threads, Zuckerberg also shared some new milestones for Threads, saying that text-based app now has 350 million monthly active users and that time spent on the platform has increased 35 percent over the last six months thanks to improvements to the company’s recommendations systems.

Later in the call, Meta’s CFO Susan Li shared that the company has also been testing its Llama model to power Threads’ recommendations and that the addition of the large language model has led to a 4 percent increase in time spent. “It remains early here, but a big focus this year will be on exploring how we can deploy this for other content types, including photos and videos,” she said.