Alphabet is doubling its capital expenditure to a staggering $180 billion in 2026 — earnings suggest that the company's AI investments may be paying off
Source: Tom’s Hardware
*Image credit: Getty / Kevin Dietsch*
During [Google](https://www.tomshardware.com/tag/google)’s Q4 earnings call last week, the company announced that it expects to **double its 2025 capital‑expenditure forecast to $185 billion**, as reported by *Reuters*【1】.
This figure is roughly **$70 billion above analysts’ expectations**. While the news sent Alphabet’s stock down 3 %, it remains one of the few AI‑focused firms whose shares have **continued to rise over the past six months**【2】.
> *“This isn’t the kind of wanton spending that the early AI‑infrastructure deals of 2025 felt like,”* the company noted.
> *“Google has real data, and to an extent, real revenue to back its words up.”*
**Key takeaways**
- **Cloud computing growth:** The cloud segment grew almost **50 % in Q4 2025**, reaching **$17.7 billion**【3】.
- **Overall revenue:** Total revenue hit **$114 billion**, up by **≈$20 billion** from the previous quarter.
---
### References
1. Reuters – *Google goes from laggard to leader in AI growth* (2026‑02‑05)
2. Reuters – *Alphabet forecasts sharp surge in capital spending* (2026‑02‑04)
3. Tom’s Hardware – *Google’s cloud computing business* (2025‑12‑XX) The Full‑Stack Approach
Image credit: Google
Unlike many other AI leaders, Google can leverage a true full‑stack approach to artificial intelligence:
| Component | Details |
|---|---|
| Software | Proprietary AI models (e.g., Gemini). |
| Hardware | In‑house Tensor Processing Units (TPU) – reducing reliance on Nvidia and other accelerator makers. Google also purchases Nvidia GPUs and will receive the first Vera Rubin GPUs later this year. |
| Infrastructure | A global network of data centers that power Google Cloud services for both consumers and enterprises. |
| Product Integration | Gemini is embedded in the Google One subscription, the Google Workspace suite (e.g., Docs), and is slated to become the default AI on every Android smartphone once the Apple partnership is finalized. |
| Monetisation Levers | • Bundling Gemini with existing Google service subscriptions. • Integrating Gemini into advertising platforms to unlock new revenue streams. • Enabling more sophisticated ad targeting for longer, complex search queries (as noted by Alphabet’s CBO Phillip Schindler). |
Why This Matters
- Revenue Flexibility: Google can package Gemini with its massive ecosystem, driving adoption through familiar services.
- Advertising Power: Enhanced AI‑driven ad solutions open up monetisation for queries that were previously hard to monetize.
- Capital Advantage: As reported by Reuters, Google was the only major tech firm to increase capital spending in Q4 2025, positioning it to sustain AI‑related investments throughout 2026 and beyond.
“Google’s full‑stack strategy gives it a unique set of levers to pull when it comes to monetising AI,” – Phillip Schindler, Chief Business Officer, Alphabet.
Outlook
With its extensive hardware, software, and cloud infrastructure, Google is poised to maintain incredible momentum in AI spending through 2025‑2026, making it an outlier among competitors. The scale of its planned 2026 capital expenditure could be difficult for even the largest rivals to match.
Even Google Needs Growth
During the earnings call, Alphabet’s finance chief Anat Ashkenazi told analysts that Google was facing a cloud‑computing backlog of $240 billion—commitments the company has pledged to meet but does not yet have the capacity to handle.
The majority of Google’s planned 2026 expenditure will be spent on:
- Reducing this backlog
- Expanding AI‑compute power for Google DeepMind
Ashkenazi added that the 2025 investment was focused on technical infrastructure such as servers, data‑centers, and networking equipment.
“We’ve been supply‑constrained, even as we’ve been ramping up our capacity.”
— Sundar Pichai, CEO of Google, recent earnings call
AI‑Infrastructure Outlook
Google’s head of AI Infrastructure, Amin Vahdat, told staff in November that the company would need to double its AI‑serving capacity every six months to keep up with demand on its cloud‑computing divisions. Even when factoring in the planned investment and efficiency gains from the latest models, that is a tall order.
“We must double AI serving capacity every six months to meet demand.”
— Amin Vahdat, internal memo (Nov 2025)
Source: CNBC article
The Growing Backlog
Despite the existing backlog, Google continues to take on new capacity orders:
| Period | Reported Cloud‑Computing Backlog |
|---|---|
| Q3 2025 | $155 billion |
| Current | $240 billion (as cited by Ashkenazi) |
That represents more than $100 billion in unfulfilled orders accumulated over just a few months. Spending its way out of this “successful hole” may be Google’s only viable path forward.
The Bottom Line
Google may be leading the AI‑infrastructure race, but there is no clear finish line. Even with genuine revenue coming from its AI divisions—and plans to expand that revenue—the spending required to retain momentum is eye‑watering. When multiple trillion‑dollar companies are chasing their own tails, the long‑term outcome remains uncertain.
Jon Martindale is a contributing writer for Tom’s Hardware. For the past 20 years he has covered PC components, emerging technologies, and the latest software advances, offering unique insights into today’s most exciting technology trends.