Central banks and investment firms have warned that if AI fails to meet investors’ expectations, there will be a sharp economic correction. Now, Deutsche Bank says 2026 will be the most challenging year for the technology yet. “AI will survive,” investment bank analysts Adrian Cox and Stéphane Avredan wrote in a Jan. 20 memo titled “The Honeymoon Is Over.” But they said the red-hot industry would be hit by three themes in particular: disillusionment, confusion and mistrust. “This is important because investment and optimism in AI is driving the global economy and accounted for much of the U.S. economic and profit growth last year,” they said. Here’s what Deutsche Bank is looking forward to in the year ahead. AI’s impact will be limited Cox and Avldan write that AI will be transformative, but it is not yet. The technology’s benefits are “more pronounced for Silicon Valley and smart early adopters” who tend to take on personal projects, they said, rather than the “average CEO” looking to improve revenue or improve operations. Most companies also don’t have the data needed to make it work at scale, Cox and Avldhan added. But AI tools for coding have advanced “rapidly.” They said that while AI agents that can complete tasks independently have been a hot topic lately, the complexities of integrating AI agents into workflows are often “ignored.” “For most people, this feels less like switching from a horse to a tractor and more like upgrading to a more comfortable saddle,” the analyst said. Bottlenecks are many There are many bottlenecks slowing down the AI race, including computing power, energy, and talent. “AI relies on the most complex supply chain in history, where any one of hundreds of thousands of components can derail the process,” Cox and Avldun wrote. Memory has become a particular crisis point as workloads shift from training models to the use of AI tools, with analysts calling it an “everyday concern” that has “temporarily distracted from the more fundamental problem of powering data centers.” At the same time, demand shows little sign of tapering off. While hyperscalers like Amazon, Microsoft, and Google pour billions into infrastructure to fuel their ambitions, small businesses are rising to take on Big Tech. The “race to secure sovereign AI” is also fueling demand outside the U.S., the analysts wrote. AWS just launched its sovereign cloud service in Europe, and Saudi Arabia is considering establishing a data embassy. “Anxiety about AI will go from low noise to loud noise this year,” Cox and Avredan wrote. “This will be reflected in lawsuits over everything from copyright to privacy to data center locations to protecting young people from chatbots that encourage self-harm and worse.” Employee turnover and abuse of chatbots are also major concerns, he said. But analysts are skeptical that AI is to blame for all of the company layoffs it causes. “AI redundancy cleaning will be a key feature in 2026,” they write. Market watchers may also be concerned about the volatile geopolitical backdrop of the AI race between the US and China. DeepSeek “showed that necessity became the mother of invention and that more value can be extracted from second-rate chips at lower costs than previously thought possible,” the analysts wrote, adding that “attempts to monopolize global standards will likely escalate.”
