- Meta invested heavily, launched proprietary Muse Spark, departing open-source.
- Muse Spark powers Meta apps, but trails top AI models.
- Developers and investors remain skeptical about Meta’s AI future.
A year after Mark Zuckerberg spent more than $14 billion to bring Alexandr Wang and a team of Scale AI engineers into Meta, the social media company has its first proprietary frontier AI model. Muse Spark, released in April this year, marks a significant shift from Meta’s open-source roots.
But the model has not yet proven itself as a serious rival to Claude, Gemini or ChatGPT, and investors remain sceptical about whether the company’s expensive AI push will actually pay off anytime soon.
Why Did Meta Walk Away From Open Source?
For years, Meta’s AI identity was anchored in Llama, its family of open-weight models that made the company one of the more accessible players in the AI space. That reputation took a hit in April last year, when Llama 4 failed to generate meaningful enthusiasm among developers, pushing Zuckerberg to rethink the company’s direction entirely.
Two months later, he announced a $14.3 billion deal to acquire roughly half of Scale AI and bring Wang on board as Meta’s Chief AI Officer, along with his most senior engineers.
According to Thomas Randall, an analyst at Info-Tech Research Group, Muse Spark was designed to work directly within Meta’s core applications, including Facebook, Instagram and WhatsApp, as well as AI-powered hardware like the Ray-Ban Meta glasses. It also powers the standalone Meta AI app and website.
ALSO READ: That Peace Sign In Your Selfie Could Let Someone Steal Your Fingerprints
“There’ll be a lot of these frontier model providers that will fundamentally change in lots of different ways, and Meta needs to have a consistent, reliable proprietary model that they themselves own,” Randall said. He added that Meta would be “lost” if Zuckerberg had not opened his wallet for Wang and other high-profile AI hires, describing the move as a “strategic rebuild” for the company.
Wang has since offered a careful framing of Meta’s open-source commitments, saying the company will continue releasing models it judges “fit and safe” to publish, while keeping frontier work locked down. The decision to keep Muse Spark proprietary was not purely commercial either. Wang acknowledged on Bloomberg Tech that internal testing flagged safety concerns that made an open release untenable.
“It actually triggered some high-risk areas in the course of early training, particularly around bio risk, but also a number of risks were elevated,” Wang told Bloomberg. He added, “This is something I think the entire industry has seen as models improved dramatically over the past year.”
Can Meta Win Back Developers And Investors?
Despite all the repositioning, Muse Spark has not landed as a credible frontier challenger. The Financial Times reported that Meta employees asked to test the model for software development tasks have continued to prefer Anthropic’s Claude. Wang has acknowledged the model trails rivals in coding, even as it has drawn praise for visual understanding.
The developer community does not appear impressed either. “I think the AI community largely ignores Meta at this point,” CNBC quoted Rob May, chief executive of startup Neurometric, which works in token engineering. May described Muse Spark as a “yawn” among the AI community, largely because the technology is not widely accessible.
On the financial side, Meta reported 33% revenue growth in the first quarter of 2025, the fastest rate since 2021. Yet its stock has fallen 18 per cent over the past 12 months, making it the worst performer among megacap tech companies alongside Microsoft. The Wall Street Journal reported that 97.6 per cent of Meta’s 2025 revenue came from advertising, while the company’s planned AI capital expenditure this year is steeper relative to its size than that of Google, Microsoft or Amazon.
ALSO READ: Meet Sarvam, The Indian AI Startup That Just Joined The Billion Dollar Club
Analysts at Truist Securities have pegged the subscription opportunity at as much as $20 billion annually by 2030, while Deutsche Bank has estimated $15.6 billion for next year alone. Those are large numbers for a company that did not clear $5 billion in non-advertising revenue last year.
Ralph Schackart, an analyst at William Blair, put it plainly: “Meta needs to provide more proof points of both adoption and commercialisation.”
Wang has described Muse Spark as an “appetiser” for larger models still in the pipeline. Whether that promise holds is something the market is clearly waiting to see.
