Klarna is changing how it handles customer support. The company is moving to a gig-based hybrid model that mixes freelance workers with artificial intelligence. The shift puts a spotlight on a familiar tension in business: how to cut costs without hurting service quality.
Why the model matters
Customer service is expensive. Full-time staff require salaries, benefits, and ongoing training. Klarna's new approach uses gig workers — people hired on short-term or part-time contracts — and AI systems to handle inquiries. That could lower expenses. But it also raises questions. Can a part-time worker know the product as well as a full-timer? Can a chatbot handle a frustrated customer?
The role of AI
AI can handle simple tasks fast. It doesn't take breaks, and it can juggle dozens of conversations at once. For Klarna, that means routine requests — like checking an order status or resetting a password — can be automated. That frees up human workers for trickier problems. But AI still struggles with nuance. It can't read tone or pick up on subtle cues. When a customer is upset, a bot might give a textbook answer that makes things worse.
The gig piece
Gig workers offer flexibility. Klarna can scale up during busy periods and pull back when things slow down. That's cheaper than keeping a fixed team. But gig workers don't always get the same training or investment. Turnover can be high. Consistency suffers. Klarna will need to figure out how to keep quality up when the workforce is fluid.
The hybrid model tries to get the best of both worlds. AI handles the volume. Gig workers handle the complexity. But the two have to work together smoothly. A customer who starts with a chatbot and then gets transferred to a human shouldn't have to repeat everything. That kind of integration is hard to pull off.
Klarna isn't the first company to try this. Many businesses are experimenting with AI and flexible labor to trim costs. The difference is in execution. If Klarna can make the model work — keeping customers happy while spending less — it could set a template for others. If it fails, the backlash could be quick. People notice when service slips.
The company hasn't announced a timeline for when the new model will be fully in place. For now, the move signals a broader shift: the old way of staffing customer service is giving way to something leaner, more automated, and less permanent. The question is whether that something can also be good.
