Klarna’s latest move to bring back some humans for customer experience after their ambitious AI...

It’s progress.

1 min read LinkedIn

It’s progress.

Too many frame this as Human vs AI.

The real transformation happens when it’s Human + AI.

We’ve implemented AI customer service agents for multiple businesses, and here’s the truth: AI can handle the bulk of common requests with superhuman speed and consistency. But when the customer experience really matters, human touch and empathy still lead.

The key isn’t to choose one or the other, it’s to design a seamless handover mechanism. Here are some examples from our experience.

AI can confidently handle:

  1. Collecting additional information from enquirers

  2. Qualifying leads

  3. Checking order status

  4. Providing business hours/location

  5. Answering FAQ-type questions

  6. Processing refund requests with standard policy

  7. Updating contact or shipping info

  8. Booking/rescheduling appointments

  9. Recommending products based on preferences

10; Following up on common ticket issues

But humans should step in when:

  1. The customer is clearly frustrated or angry

  2. There’s a complex billing or legal dispute

  3. Policies need exceptions or discretion

  4. The AI fails to understand intent twice

  5. There’s a high-value customer at risk of churn

  6. There’s a potential PR/social media escalation

  7. The customer expresses vulnerability or distress

  8. A personalised upsell/cross-sell opportunity arises

  9. The issue spans multiple departments

  10. The AI escalates uncertainty (low confidence threshold)

AI is a tool. People are the competitive edge.

Let’s stop chasing 100% automation and start building AI-augmented customer service.

Kudos to Sebastian Siemiatkowski for his open sharing.

https://lnkd.in/grB4j_Vs

#AI #CustomerExperience #CX #HumanInTheLoop #Klarna #AITransformation #FutureOfWork

Enjoyed this? Subscribe for more.

Practical insights on AI, growth, and independent learning. No spam.

More in AI Agents

Has Cursor Gotten Worse Over the Last 4 Months?

Has Cursor Gotten Worse Over the Last 4 Months?

When I first started using Cursor, I was blown away. With a single prompt, it generated clean, multi-file codes that mirrored exactly how I would have writte...

Vibe CodingAI AgentsIndependent Learning
OpenClaw vs Claude vs Hermes vs NanoClaw

OpenClaw vs Claude vs Hermes vs NanoClaw

If you have used Claude Code heavily, you saw it immediately: OpenClaw is a chat wrapper on top of CLI agents like Claude Code and Codex. So I dismissed it a...

AI Agents
Does Qwen 3.5 live up to the hype?

Does Qwen 3.5 live up to the hype?

I tested 9 local LLMs on a Claude Code skill I actually use every day. Not a coding benchmark. A real multi-step agentic task described in natural language a...

AI AgentsVibe Coding
Building AI agent skills is the most important thing you can learn right now. There are 4 levels of skill complexity.

Building AI agent skills is the most important thing you can learn right now. There are 4 levels of skill complexity.

If you use an AI agent out of the box, you are running the same model as everyone else. It gives you average output. It does not know your work, your standar...

AI Agents
Don't believe the BS that you can use Claude Code for free.

Don't believe the BS that you can use Claude Code for free.

Ollama recently made their API compatible with Claude Code. Many creators quickly jumped on the opportunity to farm engagement with the hook: "You can now u...

Vibe CodingAI AgentsAI Strategy
OpenClaw Is One of the Most Expensive Ways to Do AI Automation

OpenClaw Is One of the Most Expensive Ways to Do AI Automation

Not because the software is expensive. It is free, MIT open source. The problem is the tokens it burns to do anything. And the hype is what has turned it int...

AI Agents