The momentum of AI trading is not slowing down. Then it's Meta's turn again.
The social media giant will acquire Singapore-based artificial intelligence startup Manas, the companies announced on Monday.
Manus made headlines in March when it previewed an AI agent that could autonomously perform tasks like reviewing resumes and analyzing stock prices. Manas was founded in China but moved to Singapore in mid-2025. Meta paid Mr. Manus more than $2 billion, the Wall Street Journal reported.
The deal is the latest in a flurry of red-hot AI investments and acquisitions this year, including Meta's $14 billion investment in Scale AI in June. From providing instant AI revenue streams to an edge in AI agents, here's why acquiring Manas could give Meta a much-needed boost in the AI race.
1. You can generate revenue quickly
Manas announced in December that the company had processed more than 147 trillion text tokens and said it had “millions” of users. It also claims annual recurring revenue of over $100 million, achieving both milestones within eight months of launch.
These numbers show that Meta is gaining startups with a built-in audience of paying users. Meta's business model to date has primarily revolved around developing free products and monetizing them by collecting user data and targeted advertising. Manus offers a free tier for basic tasks, but charges users up to $200 per month for the pro tier.
“This acquisition gives Meta a functioning business with paying customers, meaningful revenue and an infrastructure that is already proven at scale,” Marcy Grundy, company profile analyst at research firm GlobalData, said in a note.
Meta said in its announcement that it plans to continue selling Manus services individually, while also integrating Manus' technology into existing platforms such as Facebook, Instagram, and WhatsApp. Mehta did not elaborate on what or how he would do so. Meta has spent billions of dollars building an in-house AI team and developing what it calls a “superintelligence,” but so far there has been little return. Manas could be a way for Meta to start monetizing directly from AI while continuing to ramp up internal efforts.
2. Manas is betting big on agents.
Meta has had a hard time surprising consumers and developers as much as OpenAI and Google when it comes to the power of its raw models. But as these models become increasingly commoditized, there is a growing need to show that AI is actually useful. One such method is AI agents. This is a type of software that allows you to proactively make decisions and take actions, such as creating marketing campaigns or monitoring and fixing bugs in your app.
Acquiring Manas may prove to be a smart bet based on the idea that the real value lies in the programs at the top of the model. Manas primarily uses AI models from other companies, such as Anthropic's Claude, as building blocks and layers its own software on top.
“People keep thinking that a small update from OpenAI or Google will kill many AI startups,” Yuchen Jin, CEO of AI startup Hyperbolic, wrote in an X post about Meta’s Manus deal. “But really, the AI application layer should be where the most opportunity lies.”
A Meta spokesperson did not immediately respond to questions about which models Manas would support after the acquisition.
3. Meta can leverage distribution advantages
One of Meta's strengths is that its platform is used by billions of people. This gives you distribution advantages similar to Google. The challenge is finding ways to keep them coming back.
Unlike Google with Gemini 3, Meta has yet to experience an AI breakthrough moment with its own model. Combining Manas' “universal agent” with Meta's distribution channels presents a new challenge for social media companies. Especially since CEO Mark Zuckerberg has acknowledged that Facebook has transitioned from a place for friends to view each other's content to a “pervasive discovery and entertainment space.”
“Manas offers an off-the-shelf, high-margin software layer that can be sold directly and integrated into Meta's consumer and enterprise products,” GlobalData's Grundy said.
