The ink barely dried on Meta's $2 billion acquisition of AI startup Manus before customers started heading for the exits. Spooked by concerns about how the social media giant handles data, some Manus users have already abandoned the platform for competitors—a stark signal that Meta's enterprise ambitions keep running into the same problem: trust. As the company tries to establish itself as a serious AI player alongside OpenAI and Google, it's discovering that a $2 billion price tag can't fix a reputation problem.
When Meta announced its $2 billion acquisition of Manus at the end of December, executives were clear about their ambitions: take this Singapore-based AI agent startup and scale it across the company's consumer and enterprise products. The vision sounded straightforward enough. But less than a month later, some of Manus's most loyal customers are already gone.
Seth Dobrin, co-founder and CEO of Arya Labs, was actually a fan of Manus before the deal closed. His company uses what's called world models—AI systems that can simulate and predict future outcomes—and he found Manus's general-purpose AI agents to be the perfect tool. Then Meta bought it. "I'm legitimately sad that this has happened," Dobrin told CNBC. "I do not agree with a lot of Meta's practices around data and how they essentially weaponize people's personal data against them."
It's not just Dobrin. Karl Yeh, co-founder of consulting firm 0260.AI, which advises startups on integrating AI tools, stopped using Manus entirely and recommended his clients do the same. The concern is simple but potentially fatal: if Meta absorbs Manus, won't Meta's data policies eventually apply to the platform? "Will the data policies of Meta apply to Manus? I would assume it will eventually," Yeh said. "That was the concern we had and why we stopped recommending it to our clients."
This is the customer exodus that nobody talks about during acquisition announcements. Manus claimed it had reached millions of paying customers with a revenue run rate exceeding $125 million when the deal closed, according to the company's blog post. The startup had built real momentum by offering AI agents that could handle complex tasks like market research, coding, and data analysis. Then one of the world's largest tech companies bought it, and some of those customers immediately looked elsewhere.
Flo Crivello, CEO of Lindy—a direct Manus competitor—actually saw a silver lining in the news. "We think there was sort of a halo effect from the announcement," he said. "It raised awareness about this category of software and people started researching it." But Crivello, who spent nearly five years at Uber working on acquisitions, thinks Meta's real play here is smaller than it appears. Manus isn't being bought to capture the enterprise software market. It's being positioned to serve small businesses and independent contractors—a move that makes strategic sense given that small businesses have long been crucial to Meta's ad revenue machine.
The problem is that customers don't know if Manus will actually remain independent as promised. "We don't know where Manus is going to fit into Meta's AI roadmap," Yeh said. "We're not sure if Manus is going to still remain a separate company even though they said it would." When companies can't commit to a product's future, customers flee. Yeh moved his company to Genspark and advised others to do the same.
What's happening with Manus is symptomatic of Meta's larger struggle in the enterprise world. The company gets nearly all of its $200 billion in annualized revenue from advertising, and every attempt to diversify into enterprise software has collapsed. Meta shut down Workplace, its communication and productivity platform, in 2024 after years of underperformance. The company discontinued Portal, its video-calling device, before that. Just last week, Meta announced it's sunsetting Workrooms, its virtual reality collaboration app—another relic from the metaverse era that nobody wanted.
None of this is happening in a vacuum. Meta CEO Mark Zuckerberg promised investors that the company would double down on AI, and he's spending accordingly. The company dropped over $14 billion in June to hire Scale AI's Alexandr Wang and a handful of his top engineers, and it's planning massive infrastructure investments to compete in the AI arms race. Yet despite all that firepower, Meta can't seem to crack enterprise.
Navrina Singh, founder and CEO of governance startup Credo AI, explained why. "Among large enterprises in highly regulated sectors like healthcare and financial services, many AI deployments today are built on models from providers such as OpenAI and Anthropic," she said. Those models then run through cloud platforms operated by Microsoft or Amazon, where "trust, security, and accountability requirements are well-established and prioritized." Meta doesn't have that reputation. In regulated industries, it's a liability.
There's one exception to Meta's enterprise drought: WhatsApp for Business. When Meta acquired WhatsApp for $19 billion back in 2014, nobody expected it to become a revenue driver. But companies discovered that WhatsApp was an effective channel for customer interaction, and it's grown into a significant business. Analyst Mark Mahaney at Evercore projects WhatsApp could generate $40 billion in revenue by 2030. That's the kind of number that gets executive attention.
So where does that leave Manus? Crivello thinks Meta will figure it out eventually, just not quickly. "The way these companies think of these acquisitions, they're acquiring the company for a specific strategic reason—they just don't know precisely what the integration might look like yet," he said. "They cut a check, it's a new thing they add to the chess board and then they figure it out. And sometimes it takes them years to figure out what to do." In the meantime, customers who had no reason to doubt Manus yesterday are looking elsewhere today.
Meta's $2 billion acquisition of Manus looked like a smart tactical play—grab an AI agent startup with real customers and revenue, integrate it into Meta's sprawling product suite, and suddenly you've got enterprise credibility. But walking that path requires something Meta doesn't have in abundance: customer trust. When Dobrin says he's "sad" about the acquisition, he's not being sentimental—he's signaling that his company lost a tool it valued, and gained uncertainty in exchange. The Manus story isn't really about whether AI agents are valuable (they clearly are) or whether $2 billion was a fair price (that's for analysts to debate). It's about the gap between what Meta wants to be in enterprise and what enterprises actually want from Meta. Until that changes, every acquisition just becomes a question mark for customers trying to decide if now is the moment to leave.
