Klarna, which said in 2024 that AI was doing the work of 700 customer service agents, starts hiring remote workers after the AI approach led to “lower quality”
To that end, Sebastian Siemiatkowski, 43, is plotting a rare recruitment drive so the buy-now-pay-later company's customers …
Context & Ripple Effects
Klarna had presented its chatbot as handling two-thirds of customer-service chats and performing the equivalent work of 700 agents in its early AI customer-service rollout. It later repeated that 700-staff-equivalent claim while reporting improved financial results in the first half of 2024, making the return to hiring a material test of whether automation savings translate into durable service quality.
The company’s staffing posture has already shifted sharply: it cut jobs in 2022, then emphasized AI tools as a way to contain costs. The planned recruitment drive reframes customer support as a workflow where automation’s output still requires a human quality backstop.
First-order effects
- Klarna will add remote customer-service capacity after finding its AI-led approach produced lower-quality outcomes, restoring human handling for cases where service standards matter most.
- The company’s prior claim that AI replaced the work of 700 agents is now qualified by an operational trade-off: lower apparent labor needs did not by itself ensure an acceptable customer experience.
Second-order effects
- Klarna must assess AI on the full cost per useful resolution—not simply chat volume or staff-equivalent output—while newly hired workers may handle escalation, exception, and quality-control work around the system.
- Other consumer-finance companies using AI support as a cost-cutting lever face stronger pressure to measure quality and retention alongside automation rates, rather than adopt headline staffing-equivalence claims uncritically.
Third-order effects
- If similar reversals persist, customer-service AI is likely to settle into a hybrid operating model: automation handles repeatable interactions, while human teams remain integral for complex or sensitive cases.
- The episode strengthens a broader shift from measuring AI by labor displaced to measuring it by reliable, workflow-level outcomes; the balance will depend on whether quality controls improve faster than the cost of human oversight.
The trend: Enterprise AI adoption is moving from broad replacement narratives toward hybrid workflows evaluated on the cost and quality of completed tasks.