Zuck's $14.3B AI Hire Drops a 'Mid' Model… And Shares Mooned Anyway

By Stocks News   |   7 months ago   |   Stock Market News
Zuck's $14.3B AI Hire Drops a 'Mid' Model… And Shares Mooned Anyway

Meta-ocrity at its finest... 

Well friends… Zuck's AI department finally did something maybe-kinda worth celebrating. Meta dropped Muse Spark on Wednesday… its first model from the new Muse series, built by the freshly minted Meta Superintelligence Labs. And for a model that's being generously described as "competitive"… shares absolutely ripped 6.5%. (the bar was in hell, but still). 

Friendly reminder that this is the first drop from The $14.3 Billion Man (read: Alexandr Wang), Meta's new chief AI officer who Zuck poached from Scale AI last June. Wang, who became the world's youngest self-made billionaire at 24, got handed the keys after Zuck fired "lost confidence in" basically everyone involved in the Llama 4 benchmark scandal. You remember… the one where Meta's own AI godfather admitted the results were "fudged a little bit" on his way out the door. (sup, Yann)

So what does Muse Spark actually do? Meta's calling it "small and fast by design"… which is corporate for "it won't embarrass us this time." The model handles reasoning, math, science, and health queries, and it's rolling out across the Meta AI app, Facebook, Instagram, WhatsApp, and those Ray-Ban glasses. Originally code-named Avocado (a.k.a, another lame AF name on the same L list as "Llama"), it even has a "Contemplating" mode where AI agents reason in parallel to compete with Gemini Deep Think and GPT Pro. And a Shopping mode. Because of course.

What’s actually interesting though… is that Muse Spark is proprietary. Meaning, it’s closed source. MEANING, the same Zuck who published an entire manifesto titled "Open Source AI is the Path Forward" just shipped a locked-down model with "hope to open-source future versions." Simply put, don't hold your breath. 

As for the spending, it’s obscene. (Should we expect anything less?). Meta’s AI capex for 2026 is between $115 billion and $135 billion. Nearly double last year. For context, OpenAI and Anthropic are collectively valued over $1 trillion, Google's Gemini has 750 million monthly users, and Meta is over here celebrating that its model is "competitive." The global generative AI market is projected to hit $325 billion by 2033… and Zuck's insatiable appetite for GPU infrastructure is burning through a third of that annually.

To which I say… might be valid. Nine months ago, Meta's AI strategy was cooked. The benchmarks were fudged, developer trust cratered, and LeCun bounced… calling Wang "young and inexperienced" in the Financial Times on his way out. Wang rebuilt the stack from scratch anyway and shipped something that matches old Llama 4 for an order of magnitude less compute. Is it frontier? Nope. But mid from the ashes of a benchmark scandal? I'm here for it.

Oh, and just to pop the cherry on all of it, Meta's also experimenting with a paid API for developers… currently limited to "select partners" (read: Zuck's speed dial). But the plan is to open it up. Translation: If you can't beat Scam Altman and Dario on models, at least sell the sh*t for cheaper. Until next time, friends… 

 

At the time of publishing, Stocks.News holds positions in Meta and Google as mentioned in the article. 

 

Did you find this insightful?

Disclaimer: Information provided is for informational purposes only, not investment advice. We do not recommend buying or selling stocks. Stock price discussions are based on publicly available data. Readers should conduct their own research or consult a financial advisor before investing. Owners of this site have current positions in stocks mentioned throughout the site, Please Read Full Disclaimer for details Here https://app.stocks.news/page/disclaimer