The “buy now, pay later” firm, valued at $14 billion ahead of its planned IPO, had trumpeted that its AI assistant could do the work of 700 human agents. But by mid-2025, the cracks were too big to ignore. Customers grew frustrated, call volumes piled up, and the company quietly began redeploying employees from across departments into call centers.
The Human Touch Returns
Siemiatkowski, who just months ago bragged about not hiring a human in a year, now admits the experiment went too far. In an interview with Bloomberg, he confessed that over-prioritizing cost “ended up lowering quality” and pledged that “investing in the quality of the human support is the way of the future for us.”
Some Klarna employees placed in the so-called “talent pool” — a holding area for staff whose jobs were eliminated — are now being reassigned to customer support. Others describe the shift as a quiet layoff strategy. According to Business Insider, even engineers and marketing professionals have been told their skills are no longer a priority and redirected to frontline service roles.
A Broader AI Reality Check
Klarna’s troubles mirror a larger trend in corporate AI adoption. A recent MIT report, The GenAI Divide: State of AI in Business 2025, found that 95 percent of attempts to integrate generative AI fail to produce financial returns. Companies like McDonald’s and Starbucks have also rolled back automation efforts after high-profile customer service mishaps.
The problem, experts say, lies in overpromising. While AI can assist with efficiency, it often struggles with nuance. Kathy Ross, a senior director at Gartner, noted in a recent survey that “the human touch remains irreplaceable in many interactions.”
Engineers on the Phones
For Klarna, the stakes are particularly high. The firm not only faces a 17 percent default rate on its microloans, as reported by Futurism, but now also a credibility crisis over its AI-first strategy. Analysts suggest that forcing highly trained engineers into call centers could signal disarray just as Klarna prepares for its long-anticipated stock market debut. Investors will be watching closely to see if the company can rebuild trust in customer service while convincing markets it still has a scalable growth model.
The AI Bubble Question
The Klarna saga underscores growing fears of an AI bubble. Despite record investments — more than $44 billion poured into AI startups in just the first half of 2025 — real-world productivity gains remain elusive. As tech critic Ed Zitron quipped, many AI “agents” are little more than “trumped-up automations” that end up creating more work than they replace.
Klarna may now serve as a cautionary tale for businesses betting too heavily on automation. The company’s new mission, as Siemiatkowski put it earlier this year, is to become “the best at offering a human to speak to.” That pivot may be as much about survival as strategy.