Google gains momentum over Meta; the outcome in 2026 will depend on the capabilities of next-generation models.

```

Author: Long Yue

Source: Hard AI

In the latest earnings season, Morgan Stanley delivered its verdict on two AI giants, declaring a “winner”.

Morgan Stanley's latest report believes that Google's parent company Alphabet, with its comprehensive earnings beat and strong upward revisions in profitability, is expected to tactically outperform its rival Meta.

The financial report shows that all of Google's major revenue streams grew more than expected, prompting the firm to raise its price target from $270 to $330. Analyst Brian Nowak pointed out in the report that Google Search business grew 14.5% year-over-year, YouTube grew 15%, and Google Cloud (GCP) was “an outstanding highlight” with an annual growth rate of 34%. This strong performance led the bank to raise its EPS forecast for Google for fiscal 2026/27 by about 7% and 9%, respectively.

In contrast, Meta's performance was described as “tactically disappointing”. Although its core platforms remain fundamentally solid, fourth quarter revenue guidance failed to boost market confidence, and capital expenditures and operating costs for the “Superintelligence Team” and core products both rose. This sparked market concerns about uncertainty in return on invested capital (ROIC), leading Morgan Stanley to lower its EPS forecast for Meta for fiscal 2026/27 by 5% and 4%, respectively, and marginally reduce its price target from $850 to $820.

However, the final outcome of this race is far from decided. The landscape in the coming years will depend on who can first release the more groundbreaking next-generation AI model.

Google Cloud business booms, profit outlook remains strong

The biggest highlight of Google's results this round comes from its cloud business. The report shows that Google Cloud's backlog orders surged 46% quarter-over-quarter, increasing by about $49 billion to a total of $155 billion. Morgan Stanley speculates that this growth may, for the first time, include orders from OpenAI and Meta.

In addition, Google Cloud's growth outlook continues to improve. The report mentions that the number of contracts over $1 billion signed by Google so far this year has already surpassed the total of the previous two years, and it stands to receive an estimated incremental revenue of $5 to $8 billion from its newly announced contract with Anthropic. Based on this, Morgan Stanley raised its 2026 and 2027 revenue forecasts for Google Cloud by 9% and 13%, respectively, projecting strong growth to continue.

Meta ramps up investment, uncertainty on returns drags down expectations

In contrast to Google's optimistic outlook, Meta is grappling with a balancing act between rising investments and uncertain returns. The report points out that Meta's operating and capital expenditures are rising due to its commitment to the “Superintelligence Team”, which is the biggest driver of its year-on-year increase in capital spending for 2026.

Although core metrics such as user time spent on Facebook and Instagram remain strong, high levels of investment have raised market concerns about Meta's return on investment. The report states that this uncertainty could put pressure on, or even cap, Meta’s stock valuation. This concern over costs and future returns has led analysts to lower their profitability expectations.

Morgan Stanley stressed that, in the current environment, upward earnings revisions are crucial to driving phase-by-phase stock price performance.

In fact, we believe that earnings revisions and confidence in the AI investment return will be key to determining the stock performance and relative valuation of Google and Meta in 2026.

Next-gen models will decide: Gemini 3 vs. Llama

Although the current earnings revisions put Google ahead, the report makes it clear that what will determine the two companies’ long-term competitive positions and valuation multiples will be their next-generation AI model capabilities.

For Google, the next key catalyst is the expected release of Gemini 3 before the end of the year. The market will closely watch for next-gen capabilities in reasoning, diffusion, and agent functions, which will be an important step in consolidating its “GenAI leadership”.

For Meta, the real test will come in 2026, when its “Superintelligence Team” needs to launch a leading and unique Llama reasoning or next-gen frontier model. The report stresses that whether Meta can use this to “rebuild its innovative leadership in GenAI” will be crucial to what kind of valuation multiples the market is willing to assign it.

This article is from WeChat Official Account “Hard AI”. For more cutting-edge AI news, follow here

Disclaimer and Risk NoticeThe market has risks, investment needs caution. This article does not constitute personal investment advice, nor does it take into account any individual user’s specific investment goals, financial situation, or needs. Users should consider whether any opinions, views, or conclusions in this article suit their particular circumstances. Investments made accordingly are at their own risk. ```