SemiAnalysis: Meta AI is expected to surpass Google within half a year, forming a three-way balance with OpenAI and Anthropic

The global competitive landscape of cutting-edge AI is undergoing a profound reshaping.

A new report from SemiAnalysis, a research firm, points out that after a year of aggressive capital投入 and architecture restructuring, Meta’s Meta Superintelligence (MSL) is expected to overtake Google in rankings of frontier AI capabilities within the next six months.

Meanwhile, Anthropic, driven by the explosive growth of Claude Code in the enterprise market, has established dual leading advantages in both profitability and growth rate, and has secretly filed for an IPO in June. The balance of power in the AI industry is evolving from a duopoly standoff between Google and OpenAI into a new tri-polar order of Meta, OpenAI, and Anthropic.

On the market front, the above developments have triggered an immediate reaction. Meta’s stock price rose by about 4% that day, while Alphabet’s stock price fell by about 1%, reflecting growing market concern about Google’s position in the AI race. At the same time, Musk publicly stated on the social platform X that Anthropic is “clearly” the current leader in the AI field, adding another variable to discussions about this competitive landscape.

SemiAnalysis’s earlier in-depth financial analysis of Anthropic’s IPO shows that the company is expected to achieve more than $1 billion in GAAP EBIT profit in the third quarter of 2026, while OpenAI’s EBIT profit margin remains negative 100%. Based on this, SemiAnalysis gives a benchmark valuation for Anthropic of $6 trillion and believes Anthropic should complete its listing ahead of OpenAI to further consolidate its leading position in the frontier model space with capital advantages.

Meta: advancing on compute, data, and talent in three lines

SemiAnalysis’s core judgment is that Meta’s pace of expansion in compute scale will allow it to surpass the combined total AI compute of OpenAI and Anthropic before the end of the year.

Citing an internal memo, Reuters reports that Meta plans to invest up to $145 billion this year in AI infrastructure, deploy 7 gigawatts of compute in 2026, and double that figure to 14 gigawatts in 2027. Supporting this goal are five gigawatt-level “titan” hyperscale data center clusters built in parallel by Meta, as well as its self-developed “AI-Backbone” network architecture—which enables Meta to asynchronously expand complex training tasks across geographic distances of thousands of kilometers.

On the chip front, Reuters reports that Meta will begin mass production in September of its self-developed AI chip codenamed “Iris.” The chip is co-designed by Broadcom and manufactured by TSMC, completes vulnerability testing in just six weeks, and has already signed multi-year supply agreements with Samsung, SanDisk, and Sumitomo Electric.

On data and talent, Meta will reassign 3,000 engineers to internal reinforcement learning (RL) environment factories to build proprietary data pipelines that commercial data providers cannot replicate. Meanwhile, Meta invests $14.3 billion in Scale AI and uses this to poach top research talent from institutions including OpenAI and Anthropic on a large scale.

SemiAnalysis believes that evaluating MSL’s current benchmark test performance is “seeing the trees but not the forest.” What truly matters is the “slope rather than the intercept”—if Zuckerberg maintains the current intensity of capital investment, Google could be permanently pushed out of the top tier of the global AI hyperscalers.

Anthropic: a B2B profit model solidifies a leading advantage

If Meta’s advantage lies in compute and infrastructure, Anthropic’s moat is built on its business model and financial quality.

SemiAnalysis’s data shows that Claude Code currently accounts for more than 7% of all code submissions on GitHub. It is this product that has driven Anthropic’s ARR from $9 billion at the end of 2025 to $30 billion in a single quarter in the first quarter of 2026—up by $3 billion in January, $7 billion in February, and $11 billion in March. As of now, Anthropic’s ARR has exceeded $60 billion, and net new ARR has reached a scale of more than $10 billion per month.

Anthropic’s financial structure sharply contrasts with OpenAI’s. Around 75% to 85% of Anthropic’s revenue comes from usage-based API businesses, while in the first quarter of 2026 OpenAI still has more than 65% of its revenue coming from subscription models, with consumer subscriptions accounting for roughly 40%. SemiAnalysis estimates that OpenAI needs to serve more than 900 million free users, at an estimated cost per user per month of about $0.70, which drags down its overall gross margin by 20% to 30%.

On profitability, SemiAnalysis expects Anthropic to achieve more than $1 billion in GAAP EBIT in the third quarter of 2026 (profit margin of about 6%), while OpenAI’s EBIT profit margin remains negative 100%. Anthropic CFO Krishna Rao previously disclosed that the company’s net dollar retention (NDR) is as high as 500%, meaning that old customers who contributed $2 billion in ARR a year ago now contribute $12 billion ARR. Based on this, SemiAnalysis predicts that if Anthropic can accelerate net new ARR to $15 billion per month, its ARR by the end of 2027 could reach $300 billion, corresponding to an enterprise value of $6 trillion.

Google: from leader to the one being chased

In this three-way contest, Google’s situation is the most delicate.

SemiAnalysis states bluntly in its report that Google has “fallen back significantly” in the competition for frontier AI, and predicts that Meta will complete overtaking Google within six months. Musk also singled out Anthropic as the current AI field’s “clearly” leading player, without including Google in the discussion of the first-tier group.

SemiAnalysis’s analytical framework shows that the decisive variables in the frontier AI competition have shifted from a single factor—model capability—to a comprehensive contest of compute scale, business model, and capital acquisition ability.

Across these three dimensions, Meta is rapidly catching up, Anthropic has built a lead, and Google faces pressure from being squeezed from both directions.

Risk warning and disclaimer

        There are risks in the market; invest with caution. This article does not constitute personal investment advice, nor does it consider the specific 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. You assume responsibility for any investment made based on this.