By Igor Pejic
Meta is taking its ad machine to the next level, but legal challenges threaten its core business and its AI strategy misses a perspective.
Henry Ford is often credited with saying that if he had asked customers what they wanted, they would have asked for a faster horse. Today, investors wish for a faster one-trick-pony, companies with a deep specialization. Mark Zuckerberg has used AI to deliver it. And that is a problem.
Emarketer now projects that Meta will overtake Google in net ad revenue by the end of 2026. Though that does not include other giants in the Alphabet world like YouTube or ads via the third-party network, Meta is catching up quickly. It’s share price is up 35% in the last 12 months. The expansion and efficiency gains of its AI business are fired up by its new AI model, Muse Spark. The model boosted personalization and engagement over the last year. Impressions are up 12%. The prices for ads rose by 9%. It thereby consolidated Meta’s position as an advertising powerhouse.
Meta is not the only big and successful company with an overreliance on a single revenue stream. TSMC makes most of its revenue as a semiconductor foundry. NVIDIA became the world’s largest company by specializing in the according infrastructure. But social media is different. It is in deep trouble and the industry will sooner or later meet the fate of Big Tobacco, as I have outlined somewhere else. Governments across the world are banning or limiting social media as it is causing empirically-proven cognitive decay, political destabilization, and harm to teenager mental health.
To his credit, Mark Zuckerberg has understood for a long time that he needs to diversify his empire beyond just adding different social media platforms. The trouble is, Meta has repeatedly failed to pull it off. And instead of a learning curve going up, some of the ideas kept getting worse and more dangerous.
A long history of diversification attempts
In 2019 Meta initiated the Libra alliance, in which Facebook and a group of like-minded companies wanted to create a blockchain-based, global super-currency that would rival the dollar and other state currencies. This was Zuckerberg’s most promising idea so far, but unfortunately for Facebook the project was ill-conceived. Instead of building a powerful dollar-based stablecoin that would have immediately turned into a multi-billion-dollar business of immense strategic importance, Libra was initially pegged to a basket of currencies. This would have given Facebook and its allies the status of a supra-national monetary player that could threaten major fiat currencies. Appetites were too high, the narrative contradictory, regulators forgotten. Facebook had to abandon the initiative due to regulatory and legal pressure.
In 2021 the metaverse pivot followed. It left Facebook with a new name (Meta) and $80-100 billion of burned cash. The idea was that we would all start donning heavy (and expensive) VR headsets to spend our time in a virtual reality instead of mingling with real people. The concept made sense for Meta’s business strategy, but for nobody else. Not even its own employees could be convinced to spend time in this new utopian world.
Meta then pivoted once again, this time to smart glasses in cooperation with Ray-Ban. Instead of just collecting data from the taps on your smartphone, Meta can now survey what you see and hear with camera’s right next to your eyes. While you might have sneered at the idea of playing poker with legless avatars in the metaverse, this product is a major escalation of privacy intrusion. Above all, those surveillance spectacles pose an unacceptable threat to children. Lawsuits are filed. Regulators are gearing up to stop it. Smart glasses are the next metaverse-type fiasco in the making.
Can AI fix it?
Thus, for societies, as well as for Meta shareholders the new focus on AI is a welcome development, especially since Meta managed to produce a high-quality model in record time. But the enthusiasm can be deceiving. The AI efforts are boosting Meta’s income and profitability in a proven business that scales. That is critically important, because AI is even outshining the metaverse in terms of spending. Meta was guiding for $115–135 billion in capital expenditures in 2026 alone and corrected those figures upward in its latest earnings’ call to $169 billion, more than twice as much as in 2025. Yet while contributing to the top line, it is questionable how Meta wants to achieve a positive return on its investment. This year’s total net ad revenue stands at a projected $243.46 billion. Given that the AI improvements contribute only a small part to it, purely tweaking the ads will not be sufficient for monetization, especially not since AI investments are scaling rapidly with no end in sight.
In other words, Meta depends on its new AI model being used more widely and outside of the company. Most other tech giants have managed this transition from one-hit wonder to versatile conglomerate. Even Apple, who champions a stable hardware business, now makes a third of its money with Apple Services.
So, can Meta pull it off? Meta is at a clear disadvantage to its competitors. It doesn’t have the cloud clout of Amazon, Microsoft, and Alphabet. It also doesn’t have the fusion possibilities of xAI, which will create a significant differentiator thanks to its merger with SpaceX. It doesn’t produce its own chips like Alphabet. And it doesn’t have the same user chokepoints like Apple (iOS and App Store), Google (search, e-mail) or Amazon (e-commerce). In other words: It is difficult to find a sustainable moat for Meta’s AI outside of its social media platforms. And this makes the entire company susceptible to a collapse in social media business. Its status a s a Big Tech player is seriously threatened.
About the Author
Igor Pejic is an award-winning author, keynote speaker, and banker. His latest book Tech Money uncovers the new rules of investing in the technology age. He is a regular speaker at institutions such as the Bank of England, the Financial Post, the German Banking Association, Euromoney, and UK Finance. Pejic publishes the Substack newsletter The New Frontier.
