Back in the AI frontline? Meta jumps 15% in a single week

Meta’s stock this week recorded its strongest week since February 2024 thanks to a series of AI strategy breakthroughs, with market confidence in its AI capabilities and commercialization path recovering significantly.

On Friday this week, Meta’s shares rose 6% in a single day, with a total weekly gain of 14.8%, marking its best single-week performance since at least February 2024, and pushing its year-to-date return back into positive territory to about 1.4%.

The last time Meta saw similarly strong performance over a single week was in February 2024, when investors responded positively to early results from the company’s “year of efficiency” cost-cutting plan—an initiative aimed at recovering the negative image caused by its previous major bet on the metaverse through financial discipline.

This stock-price breakout suggests Meta may be gradually shedding the market label of “AI laggard,” which would also open room for it to further step up its AI strategy.

Wall Street Insights noted that on July 9, Meta rolled out its flagship model Muse Spark 1.1, which has already outperformed Google’s Gemini model across multiple test categories including Agent capabilities, programming, and multimodality.

Meanwhile, Reuters reported that Meta is advancing a production ramp-up plan for its in-house chips and significantly expanding its compute infrastructure. On that basis, Deutsche Bank analyst Benjamin Black raised his estimate for Meta’s potential incremental revenue from third-party cloud services from $17 billion to $24 billion.

Research firm SemiAnalysis released a report forecasting that Meta’s Meta Superintelligence (MSL) could surpass Google in rankings of frontier AI capabilities within the next six months, shifting the AI competitive landscape from a two-horse dominance by Google and OpenAI to a three-way standoff among Meta, OpenAI, and Anthropic.

Muse Spark 1.1’s low-price strategy spells a price war

The Muse Spark 1.1 model launched by Meta this week is its first commercially available model with near-frontier-level agent programming capabilities, and it is also equipped with a paid API interface.

On Thursday, CEO Mark Zuckerberg posted on the X platform emphasizing that the model is priced “very affordably,” sparking widespread market speculation that Meta would proactively launch an AI inference price war to pressure competitors.

The Meta Model API provides a $20 free credit per account, charges on a pay-as-you-go basis, with input priced at $1.25 per million tokens and output priced at $4.25 per million tokens.

Richard Windsor, founder of Radio Free Mobile, pointed out in his Friday research note that the release of Muse Spark 1.1 confirms recent reports that Meta plans to launch a new business selling compute capacity. Windsor wrote:

There is growing evidence that, given the currently attractive return rates, Meta will roll out a new business to sell compute capacity to third parties.

He further noted that Muse Spark is already approaching top-tier models in AI programming capability, “but the price is only 25% of the latter,” making it extremely appealing to the mass market.

In-house chip development and compute expansion boost cloud revenue potential

Wall Street Insights noted that Meta has planned to start mass production of its in-house AI chip codenamed “Iris” in September this year. The chip is designed jointly with Broadcom and manufactured by TSMC, completes testing in just six weeks, and has already signed multi-year supply agreements with Samsung, SanDisk, and Sumitomo Electric.

In terms of compute scale, Meta plans to deploy 7 gigawatts of compute capacity this year and double that to 14 gigawatts in 2027.

Supporting these goals are five gigawatt-class “titan” ultra-large data center clusters that Meta is building in parallel, as well as its in-house “AI-Backbone” network architecture, which allows Meta to asynchronously scale complex training tasks across geographic distances of thousands of kilometers.

In a research note on Thursday, Deutsche Bank analyst Benjamin Black said that the above compute expansion plans imply Meta’s potential incremental revenue from third-party cloud services is about $24 billion, significantly higher than the earlier $17 billion forecast.

He also noted that Meta’s in-house chips are expected to open up a practical path to lower costs and improve efficiency for the company.

SemiAnalysis: Meta AI could surpass Google within half a year

Wall Street Insights noted that research firm SemiAnalysis believes that after a year of aggressive capital spending and architectural restructuring, MSL is expected to surpass Google in rankings of frontier AI capabilities within the next six months.

The report states that the current two-stronghorse pattern of Google and OpenAI will be rewritten into a three-way standoff among Meta, OpenAI, and Anthropic.

SemiAnalysis’s core judgment hinges on the speed of compute expansion: Meta’s growth trajectory in AI compute capacity will cause it to surpass the combined total compute of OpenAI and Anthropic before the end of the year.

According to Reuters citing an internal memo, Meta’s cap on capital expenditures for AI infrastructure this year could be as high as $145 billion.

On talent: Meta has reassigned 3,000 engineers to an internal reinforcement learning environment factory, building proprietary data supply that commercial data providers can hardly replicate. It also invested $14.3 billion in Scale AI to bring in top research talent at scale from institutions including OpenAI and Anthropic.

SemiAnalysis believes that evaluating MSL purely based on current benchmark test results is “seeing only trees but not the forest.” The truly important factor is future momentum, not the height or depth of the starting point today.

The report also indicates that if Zuckerberg maintains the current pace of capital investment, Google could be permanently excluded from the first tier of global AI super-scale players.

Risk Disclosure and Disclaimer

        There are risks in the market; invest with caution. This article does not constitute personal investment advice, and it does not consider any special investment objectives, financial conditions, or needs of individual users. Users should consider whether any opinions, viewpoints, or conclusions in this article align with their specific circumstances. Invest accordingly at your own risk.