Edition #13 - Humans in the Loop

AI-first strategy only works when humans stay in control

By Mykel Salomon ·

Opening Reflection

“Humans in the loop” gets thrown around like a safety blanket. But if we don’t define it, it becomes a slogan, and slogans don’t protect customers, agents, or brands.

AI is here to stay. It will accelerate. It will automate. It will compress time. And in support, that speed can either become a superpower… or a trust disaster.

Humans in the loop is the difference.

What “Humans in the Loop” Actually Means

At its simplest: AI executes at scale. Humans retain judgment and accountability.

Not as a last resort. Not as “we’ll step in if it breaks.” As a designed, visible control system.

In customer support, “humans in the loop” means:

This is systems engineering at work.

🌍 In the Wild: How Real Companies Are Doing It

1) AI-first chat + human handoff

Klarna reported its AI assistant handled two-thirds of customer service chats in its first month, with faster resolution times and fewer repeat inquiries. But the follow-up lesson matters too: Klarna later re-emphasized human talent in customer service, because efficiency gains don’t automatically equal trust and quality at scale.

Takeaway: AI can absorb volume. Humans protect the relationship.

2) AI agent resolves — humans govern the edges

Intercom ’s Fin is explicitly designed with human handoff controls, including workflows that route to humans when it’s the “safest option.” Some Intercom customer stories report high resolution rates (for example, Lightspeed reporting significant AI resolution volume), but the consistent pattern is: AI starts; humans remain reachable and informed.

3) “Copilot” inside the agent workflow

Zendesk Copilot focuses on helping agents with intelligent triage, summaries, and agent assist—AI behind the scenes, humans delivering the experience. Zendesk has also shared that AI triage can save an average of ~45 seconds per ticket compared to manual triage.

4) Case summaries + knowledge creation with review

Microsoft Dynamics 365 Customer Service Copilot supports case summaries and can draft knowledge articles from resolved cases, with humans reviewing and editing before publishing.

Takeaway: HITL isn’t only frontline chat. It’s governance across the whole support lifecycle.

The AI-First Support Blueprint

Here’s what “Humans in the Loop” looks like as a real structure.

Layer 1 — Front Door AI

Channels: chat, email, web, in-app, social. AI handles: FAQs, simple workflows, retrieval from knowledge base, order/status lookups.

Guardrails:

Layer 2 — Orchestration + Retrieval

A good doesn’t “wing it.” It retrieves answers from approved sources (knowledge base, policies, product docs), then responds within constraints.

Layer 3 — Human Handoff by Design

When risk rises, the system escalates before damage happens:

And the handoff must include:

(Intercom and Zendesk both emphasize this kind of assisted workflow/handoff capability in different ways.)

Layer 4 — Copilot for Agents

AI drafts replies, summarizes threads, suggests macros/knowledge. Humans edit and send. Humans own tone. Humans own judgment.

Layer 5 — Quality + Feedback Loop

Every escalation, correction, and failure becomes training data:

This is where most companies fail: they deploy AI and skip the learning system.

How to Roll This Out Without Breaking Trust

Phase 1 — Start narrow

Pick 10–20 low-risk intents (password reset, order status, “how-to” basics). Measure deflection and customer frustration.

Phase 2 — Shadow mode

Let AI draft responses internally first. Agents approve. Agents correct. You learn safely.

Phase 3 — Confidence thresholds

If the bot is uncertain, it escalates. Period. Don’t let it guess.

Phase 4 — Escalation design

Define “always-human” categories:

Phase 5 — Human capability upgrade

Train agents to become:

AI reduces repetitive work. Humans become the difference.

The Metrics That Actually Matter

If you only chase deflection, you’ll build a fast machine that customers hate.

Track:

Zendesk’s “time saved per ticket” is real value, if it translates into better outcomes, not just faster closure.

💬 The Big Question

Where do you need humans for trust, not because AI can’t answer, but because customers need to feel heard?

🧠 A 10-minute exercise for CX & Support leaders

Write this down with your team:

  1. List your top 15 ticket types.

  2. Mark each as: AI can solve / AI can assist / Human must own.

  3. For “AI can solve,” define the confidence threshold and handoff trigger.

  4. For “Human must own,” define what AI can do behind the scenes (summary, retrieval, draft).

This becomes your HITL map.

🎙️ This Week’s Episode

Season 2 · Episode 3 The Story of Support Driven — and the Power of Community

Guest: Scott Tran | Support Driven

If “Humans in the Loop” is the blueprint for AI-enabled support, this episode is the reminder of why it matters.

Scott Tran built Support Driven from the simplest beginning: two people in a Slack. No scale. No hype. Just consistency, care, and showing up, one human at a time. And that’s the lesson that hits hardest right now.

Because in a world racing toward automation, community becomes a form of protection. Not from technology , but from what happens when technology moves faster than trust.

In this conversation, we explore:

If you’re building AI into support, don’t skip this. Because your automation will only be as strong as the human network behind it.

🎧 Listen or watch the episode:

Article content

Thought to End the Week

Automation without humans in the loop is just speed with no conscience. Build AI-first support, yes. But keep humans where trust is won, and where it’s repaired.

— Mykel