Editorial
Research Report
The 2026 State of CX Outsourcing
The Gist
- The AI layoff wave will boomerang. Gartner predicts that by 2027, half of companies that cut customer service staff due to AI will rehire—underscoring how often automation strategies overestimate capability and underestimate customer complexity.
- Cost-first AI creates a customer experience gap. When implementations focus on headcount reduction instead of solving real customer problems, organizations erode trust, overload remaining agents, and degrade service quality.
- The winning model is augmentation, not replacement. AI strengthens service when it enhances identity, intent and information—supporting human judgment rather than attempting to eliminate it.
A recent Gartner prediction should make every CX leader pause: By 2027, 50% of companies that cut customer service staff due to AI will rehire people to perform similar functions, but under different job titles.
This isn't just a workforce trend. It's a symptom of a fundamental misunderstanding about what AI can and cannot do in customer service.
Table of Contents
- The Rush to Implement AI Without Strategy
- The Real Motivation: Cost Reduction Over Customer Value
- Where AI Is Failing Customers
- The Customer Experience Gap
- What AI Actually Delivers Today
- The Coming Rehiring Cycle
- A More Strategic Approach to AI in Customer Service
- The Three Pillars of Effective Human-AI Collaboration
- The Question Every CX Leader Must Answer
The Rush to Implement AI Without Strategy
According to Intercom's 2026 Customer Service Transformation Report, 82% of senior leaders invested in AI for customer service in the last 12 months, and 87% plan to make additional investments in 2026.
But here's the critical question: How many of these investments started with identifying a specific customer problem versus calculating potential headcount reductions?
In my experience working with organizations implementing AI, I've witnessed the pattern repeatedly. A large travel organization I consulted with wanted to implement AI with no clear plan or strategy, just a mandate that "we need AI." They tasked two employees with driving experimentation. After months of effort with no meaningful results, the project was quietly shelved.
This approach — implementing AI for the sake of AI — is becoming the norm rather than the exception.
Related Article: Insights, Infrastructure and AI: What Real CX Maturity Looks Like
The Real Motivation: Cost Reduction Over Customer Value
The uncomfortable truth is that most AI implementations in customer service are driven by a single metric: how many agents can we eliminate?
This inside-out perspective focuses on organizational efficiency rather than customer outcomes. It asks, "How can we reduce costs?" rather than "How can we solve customer problems more effectively?"
Kathy Ross, senior director analyst in Gartner's Customer Service & Support practice, noted: "While AI-driven layoffs have captured attention, the reality is more complex. Most recent workforce reductions were influenced by broader economic conditions rather than automation alone. As organizations encounter the limits of AI and rising customer expectations, they will need to reinvest in human talent to sustain service quality and growth."
Where AI Is Failing Customers
The limitations of current AI implementations are becoming increasingly visible through high-profile failures:
- Air Canada's Chatbot Liability: The airline was held legally liable after its chatbot provided incorrect information to a passenger, demonstrating that organizations remain accountable for AI-generated advice.
- Cursor's Fabricated Policies: The AI chatbot created entirely fictional policies, misleading users and eroding trust in automated support channels.
- Klarna's Reversal: After initially replacing customer service staff with AI, Klarna reversed course and resumed hiring human staff, acknowledging the limitations of its automation strategy.
These aren't isolated incidents. They represent the fundamental challenge of implementing AI without understanding its constraints.
The Customer Experience Gap
Consider the typical customer journey in an AI-first support environment:
- Customer visits the website
- Reviews the FAQ section
- Interacts with a chatbot
- Speaks with voice AI
- Finally reaches a human agent, frustrated and expecting expert-level assistance
At each automated touchpoint, the customer's expectations rise. When they finally reach a human, they expect that person to solve anything, because every other channel has failed them.
But here's the problem: Organizations have often eliminated their most experienced problem-solvers during AI implementation, retaining only those who can follow predefined scripts and workflows.
What AI Actually Delivers Today
After extensive research and consultation with organizations implementing AI in customer service, I have yet to find a compelling case where AI or Agentic AI enabled customers to accomplish something they genuinely couldn't do before.
In most implementations, AI functions as a more interactive FAQ or knowledge base. It provides quick access to documented information, but struggles with:
- Complex, multi-faceted problems requiring judgment
- Situations requiring empathy and emotional intelligence
- Novel issues not covered in training data
- Contexts where customers need to be heard, not just answered
This is not innovation. It's automation of the basics, and there's a meaningful difference.
Related Article: OpenAI CEO Sam Altman Says AI in Customer Support Is 'Doing Great'
The Coming Rehiring Cycle
Gartner's prediction that 50% of companies will rehire by 2027 reflects an inevitable correction. As AI's limitations become apparent and customer satisfaction metrics decline, organizations will need to bring back human expertise.
But they'll do it under different job titles: Customer Service Representatives become Solution Consultants. Support Agents transform into Trusted Advisors or Product Specialists.
Why the title changes? Because the AI investment must be justified, even when it has demonstrably failed to deliver on its promises. Changing titles allows organizations to claim the AI implementation succeeded while simultaneously acknowledging they still need human intelligence.
These new roles will focus less on transactional problem-solving and more on ensuring customers extract maximum value from products and services. It's a shift from reactive support to proactive guidance, but it still requires the human judgment and relationship-building capabilities that AI cannot replicate.
A More Strategic Approach to AI in Customer Service
Four shifts CX leaders should make when implementing AI in customer service.
| Strategic Principle | What It Means in Practice |
|---|---|
| Start with the customer problem | Before any AI implementation, clearly articulate what specific customer challenge you're addressing. If the answer is primarily about internal efficiency rather than customer value, reconsider your approach. |
| Focus on augmentation, not replacement | Identify use cases where AI enhances human capability rather than attempting to eliminate it. This includes providing agents with relevant context before interactions, surfacing appropriate knowledge articles during conversations, handling genuinely routine transactional requests and analyzing patterns across interactions to identify systemic issues. |
| Preserve human judgment for complexity | Recognize that the most valuable customer interactions — those involving trust, empathy and complex problem-solving — require human intelligence that current AI cannot match. |
| Measure customer outcomes, not just efficiency | Success metrics should focus on customer experience, problem resolution quality and relationship strength, not solely on cost per contact or deflection rates. |
The Three Pillars of Effective Human-AI Collaboration
The foundational elements that determine whether AI strengthens or weakens agent performance.
| Pillar | Description | How AI Enhances It |
|---|---|---|
| Identity | Agents must know who the customer is, including their history, value and context. | AI aggregates data from disparate systems to present unified customer profiles in real time. |
| Intent | Understanding why the customer is reaching out and what they're trying to accomplish. | AI predicts needs, detects sentiment and surfaces likely goals before or during the interaction. |
| Information | Having immediate access to the knowledge required to solve the customer's specific issue. | AI delivers relevant articles, policy guidance and contextual recommendations during conversations. |
Webinar
Feb
25
Content Strategy Leaders Live: Scaling for Speed, Complexity and AI in High Tech
A candid roundtable on how high-tech leaders are rethinking content at scale.
Register
Webinar
Feb
26
Why Customer Insights Never Turn Into Action
The gap between insight and action is solvable...once you understand how scattered knowledge can wreak havoc on execution.
Register
Webinar
Feb
26
Do More with Less: Modernizing the Cloud Contact Center for 2026
Learn how to leverage cloud platforms without adding a single hire to personalize every customer interaction.
Register
Webinar
Mar
25
Content Leaders Collective: Navigating Content Decisions at Scale
Discover how content leaders are modernizing content operations, avoiding costly missteps and preparing for scale and AI.
Register
Webinar
Mar
26
Do Learning Programs Really Work? How to Turn Education Into Engagement In Healthcare
See how leaders are using learning programs to build trust with healthcare professionals and create measurable engagement.
Register
Webinar
On demand
Cut the Noise: Deploying AI That Actually Moves the Needle
Learn how to turn AI experimentation into concrete revenue operations.
Watch Now
AI can significantly enhance all three pillars by aggregating data from disparate systems, predicting customer needs and delivering relevant information in real time. But the human agent must still synthesize this information, apply judgment and make the final decision.
The Question Every CX Leader Must Answer
Before implementing AI in customer service, ask yourself: What specific customer problem does this solve that we cannot solve effectively today?
If your honest answer is "reduce headcount," "cut costs," or "we need to be seen as innovative," you're likely setting yourself up to join Gartner's predicted 50% who will need to rehire.
The organizations that will succeed with AI in customer service are those that view it as a tool to enhance human capability, not replace human judgment. They're solving customer problems, not just budget problems.
The choice is clear: Implement AI thoughtfully to strengthen your customer relationships, or implement it hastily and watch those relationships (and your workforce) deteriorate, only to rebuild what you dismantled.
Which path will you choose?
Learn how you can join our contributor community.