مقال

Big tech’s 2025 AI capex race: Amazon extends its lead

November 26, 2025
مقال

Big tech’s 2025 AI capex race: Amazon extends its lead

November 26, 2025
مقال

Big tech’s 2025 AI capex race: Amazon extends its lead

November 26, 2025

The scale of investment heading into 2025 is unlike anything the industry has seen. Amazon, Microsoft, Alphabet and Meta together are guiding toward $360–$400 billion in capital expenditure - an increase of roughly 60% year-on-year. Almost all of this spending is earmarked for AI infrastructure: hyperscale data centres, custom silicon, networking, and massive GPU and accelerator clusters designed for training frontier models.

On 24 November 2025, BNP Paribas Exane initiated coverage of Amazon with an Outperform rating and a $320 price target - the highest among major brokers. Based on Amazon’s 26 November close of around $230, this implies nearly 39% upside, reflecting heightened confidence in Amazon’s aggressive AI build-out.

2025 capex guidance - The big four

Company 2025 Capex Guidance Primary AI Focus Areas
Amazon >$125 bn (raised multiple times in 2025) AWS hyperscale clusters, Trainium/Inferentia chips, sovereign & government clouds
Microsoft $80–121 bn (FY ending Jun-26) Azure expansion, OpenAI infrastructure, enterprise GenAI
Alphabet ~$91 bn (raised from $85 bn) Google Cloud TPUs, catching up on historical capacity shortages
Meta $70–72 bn Llama models, AI-driven advertising, massive single-site data centers

Why BNP Paribas Exane sees Amazon differently

Analysts at BNP Paribas Exane argue that concerns about Amazon’s pace of AI investment are misplaced. Based on management commentary, Amazon’s 2025 capital expenditures (capex) are expected to reach nearly $125 billion, with the possibility of increasing even higher in 2026. Almost all of this is channelled into AI infrastructure - from new AWS availability zones to networking improvements and the next generation of Amazon-designed accelerators.

The brokerage highlights three factors that distinguish Amazon from its peers in this capital expenditure cycle. First is vertical integration: Amazon’s Trainium and Inferentia chips allow AWS to manage costs more effectively and alleviate dependency on third-party GPU supply. Second, Amazon has multiple monetisation pathways. 

The same infrastructure that powers enterprise AI workloads on AWS also enhances Amazon’s advertising relevance, retail logistics, and consumer services. Third is the margin narrative: with AWS anticipated to reaccelerate into the mid-20% range and advertising growing above 20% annually, Amazon could see multi-year improvements in operating leverage - provided execution remains strong.

Key investor debates & risks

Debate / Risk Representative “Bull” Perspective Representative “Bear” Perspective
Scale of capex Large-scale AI capex is seen as necessary to secure long-term demand in cloud, AI services and advertising, with the view that current spending reflects structural growth in workloads. Some investors worry about an overbuild scenario where capacity is added faster than demand, leading to lower returns on invested capital and potential assets that are underutilised.
Timing of returns Supportive commentators expect utilisation and monetisation to ramp meaningfully through 2026–2027 as more generative AI projects move from pilot stages to full deployment, particularly in cloud and enterprise software. Sceptical views focus on free cash flow compression in the near term and uncertainty about how quickly enterprises will translate experimentation into large, recurring AI spending commitments.
Competitive positioning Proponents see Amazon’s full-stack approach (from custom chips to cloud to consumer applications) as a durable advantage relative to peers focused more on individual layers of the stack. Others point to the rapid product and ecosystem momentum at Microsoft’s Azure and Alphabet’s Gemini/Google Cloud offerings and question whether any one company can sustain a clear lead.
Macro sensitivity Some argue that cloud and AI-related spend are becoming more “infrastructure-like” and may remain relatively resilient even if consumer spending slows, especially in mission-critical workloads. There are concerns that a broader economic slowdown could still affect digital ad budgets and e-commerce volumes, which are important parts of revenue for Amazon and Meta.

Upcoming catalysts/data points

AWS re:Invent - early December 2025

Investors will watch closely for new AI services, model announcements, TPU/Trainium updates and customer case studies demonstrating deployment at production scale.

Amazon Q4 2025 results - late Jan/early Feb 2026

AWS revenue growth, segment operating income, cloud utilisation and commentary on 2026 AI capex plans will be key market-moving metrics.

Peer earnings - early 2026

Microsoft, Alphabet and Meta will provide fresh updates on capex levels, AI model adoption and how each business is balancing deep investment with free-cash-flow discipline. These read-across signals will shape expectations for whether 2026 spending remains elevated or begins to moderate.

These milestones will help clarify how quickly AI infrastructure investment is converting into revenue - and whether the current capex boom has overshot or remains justified.

Amazon technical insights

At the time of writing, Amazon (AMZN) is trading around $229, forming a modest recovery from recent lows while staying above notable support zones at $218.45 and $213. A break below these levels risks triggering sell-side pressure and accelerating downside momentum. A strong move higher, however, puts $250.15 back into focus - an area traders may treat as a profit-taking or breakout opportunity.

The RSI sits near 50, signalling neutral momentum and suggesting the market is still digesting the recent pullback. Directional conviction may not emerge until broader sentiment around AI spending turns clearer.

Source: Deriv MT5
إخلاء مسؤولية:

The performance figures quoted are not a guarantee of future performance.

الأسئلة الشائعة

How much of Amazon’s capex is explicitly AI-related?

Amazon has stated that the “vast majority” of its projected $125 billion 2025 capital expenditure is focused on AI and cloud infrastructure, including data centres, networking upgrades, and in-house silicon for AWS workloads, although it has not provided a strict numerical breakdown.

When do companies expect AI investment to become profit-accretive?

Across large tech firms, commentary suggests 2026–2027 as the period when higher data-centre utilisation and AI service adoption should begin generating more visible operating leverage. The exact timing will depend on demand, pricing and cloud workload mix.

Which public companies benefit most from the AI infrastructure boom?

Frequently mentioned beneficiaries include Nvidia, Broadcom, Micron, TSMC, Vertiv and Super Micro Computer, all of which gain from the build-out of AI chips, memory, power systems and server hardware.

المحتويات