Quarterly Outlook
Q4 Outlook for Investors: Diversify like it’s 2025 – don’t fall for déjà vu
Jacob Falkencrone
Global Head of Investment Strategy
Investment Strategist
Meta plans to cut a large share of its metaverse budget and redirect capital and talent into AI infrastructure, wearables and “superintelligence” projects.
The core ads business already monetises AI, but record capital spending raises questions about how fast new AI products will pay for themselves.
Long-term investors should treat Meta as a cash-rich, founder-led AI utility and track discipline on capital spending, Reality Labs losses and the rise of new AI revenue streams.
A few years ago, Meta rebranded the whole company around the metaverse and poured tens of billions into virtual reality headsets and digital worlds. Reality Labs, the division behind Quest devices and Horizon Worlds, has since racked up more than 60–70 billion dollars in losses, while user adoption stayed modest.
Now the script is changing. Meta is considering cutting up to about 30% of its metaverse budget from 2026, with Reality Labs expected to bear much of the reduction. Management is already shifting money from pure metaverse projects towards AI-powered glasses and wearables, where products such as the Ray-Ban Meta smart glasses show more promising traction.
The question for investors is simple: is this a long-overdue clean-up of a costly side quest, or just a rotation into another expensive, hard-to-monetise vision?
Meta is turning itself into one of the world’s largest AI factories. For 2025, management expects capital expenditure of about 60–65 billion dollars, mainly on data centres, chips and other infrastructure that power its apps and new generative AI projects. On current estimates, capex is set to rise above 100 billion dollars as soon as 2026, showing how quickly the spending ramp is accelerating.
Crucially, this spend already feeds a very profitable engine. In the third quarter of 2025, revenue grew about 26% year on year to just over 51 billion dollars, helped by AI-driven upgrades to its ad systems that better match adverts to users across Facebook, Instagram and Reels. Operating margins sat around 40%, reminding investors that the old ad machine is still very much alive.
AI ad tools such as Advantage+ now generate more than 60 billion dollars in annual run-rate revenue. For advertisers the promise is lower cost per lead and better conversion. For Meta, each ad impression becomes more valuable, cushioning past privacy hits and lifting returns on that infrastructure.
The hiring strategy matches the scale of the bet. Meta has invested heavily in Scale AI and brought its founder, Alexandr Wang, in as Chief AI Officer to run Meta Superintelligence Labs, while slimming bureaucracy in older AI teams and targeting what Wang calls “hotshot” AI scientists.
Markets broadly like the AI story, but they still worry about the bill. Meta’s shares have bounced on news of metaverse cost cuts, yet remain sensitive to any sign that AI spending will stay “higher for longer.”
For now, most of the cash still comes from familiar places. Meta’s AI keeps people scrolling, surfaces more engaging content and serves more relevant ads. Short-form video on Reels has moved from being a drag on profits to pulling its weight in monetisation, helped by better recommendation models and ad formats.
New AI products are earlier in their journey. Meta has rolled out chatbots, AI features inside WhatsApp and Messenger, and early versions of personal assistants that live inside its apps and devices. The long-term vision is that AI agents help users shop, manage tasks, create content and interact with businesses, with Meta taking a fee or ad share along the way.
AI glasses and other wearables are the wild card. Here Meta hopes that always-on AI assistants, accessed through normal-looking glasses rather than bulky headsets, can finally make its years of hardware work pay off. Trimming pure metaverse projects and leaning into AI-first wearables shows management is now more willing to follow the user than the buzzword.
The main risk is that spending stays high while monetisation lags. Building world-class AI models and data centres is expensive, and Meta is competing directly with Alphabet, Microsoft and major Chinese platforms that are all chasing similar customers.
There is also execution risk. Meta is reorganising its AI efforts, hiring star talent and seeing some long-standing research leaders move on, including its veteran AI chief Yann LeCun as the strategy shifts from open research to more commercial products. That can be healthy, but it increases the chance of missteps just as the company is betting its next decade on AI.
On top sit regulatory and political risks, from content rules to AI safety concerns and possible new taxes on big tech. And as always with Meta, investors live with a tight governance structure where Mark Zuckerberg holds firm voting control. Founder-led can be a strength, until it is not.
Meta’s story used to be simple: a very profitable social network quietly funding a very expensive metaverse dream. The new chapter is tidier. Reality Labs is shrinking, AI is moving to centre stage, and more of the budget now supports products that already touch billions of users.
For investors, the opening question still stands: is Meta moving from one money pit to another, or from a failed dream to a disciplined AI factory? The difference this time is that AI spending is already lifting ad performance and engagement. If management keeps that focus and shows clear payback on capex, Meta’s pivot could end up compounding something very old-fashioned: cash flow.
This material is marketing content and should not be regarded as investment advice. Trading financial instruments carries risks and historic performance is not a guarantee of future results.
The instrument(s) referenced in this content may be issued by a partner, from whom Saxo receives promotional fees, payment or retrocessions. While Saxo may receive compensation from these partnerships, all content is created with the aim of providing clients with valuable information and options.