Mohamed El-Erian thinks that last year's economic tailwinds will give way to more challenging conditions as the AI narrative fades

By Mohamed El-Erian*

 For global markets, 2025 was defined as much by what did not happen as by what did. The year offered a masterclass in the power of a single narrative, with massive, concentrated bets on AI masking various other unanswered questions. Yet as we move further into 2026, the AI narrative is unlikely to prove strong enough to continue overshadowing other lingering uncertainties, many of which reflect deeper structural shifts. For investors, central banks, and governments alike, the situation demands adaptation.

Many economic developments in 2025 served as a direct rebuke to the conventional wisdom. Despite dire warnings of a “lose-lose-lose” trade war, US tariffs did not trigger a stagflationary spiral or provoke as much retaliation as many expected. Even more surprising was the adaptability of China’s export engine. By rerouting trade through non-US partners, especially in Europe and Asia, China achieved a record-breaking trade surplus exceeding $1 trillion.

In the United States, consumers also proved extraordinarily resilient, especially low-income households, which continued to spend despite high prices, mounting debt, and growing anxieties about jobs and incomes. When combined with investors’ willingness to throw money at “anything AI” and to continue financing large deficits in the US and other advanced economies, this resilience fueled a robust US economic expansion and generated beneficial spillovers for the global economy. With third-quarter growth accelerating to 4.3%, the US economy crushed expectations.

Beneath the surface, however, unsettling structural trends have begun to emerge in the world’s most influential economy. We are witnessing a decoupling of US GDP growth from job creation, even in these very early stages of AI, robotics, and related technologies’ broader adoption. Moreover, a widening “K-shaped” divergence (with the richer economic cohorts doing much better than others) is turning affordability into a volatile political and social flashpoint. The increasingly important influence on policy implications is reflected in the Trump administration’s recent focus on housing and energy.

Globally, it is hard to see how other major economies, especially Europe, can continue to accept the massive volumes of Chinese exports that have been redirected their way. The repeated breakdown of international policy coordination – vividly illustrated by the US boycott of the G20 in South Africa – confirms that the “multilateral” era is in the rearview mirror, at least for now. It has been replaced by a more fragmented landscape where geoeconomics dominate. Never mind that closer coordination is needed more than ever to deal with a wide range of common challenges, both longstanding and new.

Wherever you look, the traditional factors underlying economic and commercial activity are likely to be increasingly sidelined by national-security concerns, geopolitics, and domestic political machinations – particularly in the US ahead of this year’s midterm elections. If 2025 was about ignoring the market spillovers of domestic and international politics, 2026 will be about navigating them.

The US intervention in Venezuela has further complicated this geopolitical dimension. Such a sudden, largely unexpected development is likely to have multifaceted “demonstration effects.” Not only will it discourage smaller powers like Colombia, Iran, and Cuba from challenging the US; it may also encourage larger powers like China and Russia to consolidate their own spheres of influence on the basis of “might makes right.”

Under these new circumstances, the narrative of AI dominance looks set to lose some salience. The animal spirits that drove indiscriminate, massive financing last year will increasingly be tamed by bubble fears that force investors to be more selective.

The fascination with those working on AI – building stunning new models – will be tempered by the realisation that the West is falling behind China and parts of the Middle East in working with AI. Especially now, when we are on the cusp of a robotics revolution, widespread adoption will be key. This integration gap will determine who dominates the next phase of global productivity growth, but it has yet to attract sufficient policy attention in the US and Europe.

For investors, the standard playbook will need to change. Riding a broad structural wave is no longer such an obvious and rewarding strategy. The 2026 outlook is likely to demand a more tactical, bottom-up approach. Success will require identifying “market-completion” opportunities – where infrastructure and new tools enable practical applications – and focusing more on tangible assets over speculative growth.

For their part, governments and central banks must recognise that they will no longer be saved by private-sector innovation and financing miracles. No longer will AI hype and ample funding offer a shield for policy failures. The transition will be especially challenging for the US Federal Reserve. The world’s most powerful central bank must move beyond the overly rigid, “data-dependent” approach that has turned it into more of a volatility amplifier than a source of stability. Unless the Fed follows through on long-needed reforms to counter the forces eroding its credibility and effectiveness, its political independence will remain at risk.

Across advanced economies, central-bank changes will need to be accompanied by comprehensive growth strategies. High deficits and ballooning debt will prove far less forgivable in a world where the AI narrative is no longer loud enough to drown out lagging policies. Muted growth outside the US, especially in Europe, whose modernisation (as advocated in the Draghi Report) is advancing at a snail’s pace, will further contribute to potential global instability.

No matter how you slice it, 2026 will not bring more of the same. It should be a year of recalibration – for investors and policymakers alike.

Mohamed A. El-Erian, President of Queens’ College at the University of Cambridge, is a professor at the Wharton School of the University of Pennsylvania and the author of The Only Game in Town: Central Banks, Instability, and Avoiding the Next Collapse (Random House, 2016) and a co-author (with Gordon Brown, Michael Spence, and Reid Lidow) of Permacrisis: A Plan to Fix a Fractured World (Simon & Schuster, 2023).  Copyright: Project Syndicate, 2026, published here with permission.