Two data points from Gartner landed this week, and reading them side by side tells a story that every business owner needs to hear.
First: 91% of customer service and support leaders now report pressure from executive leadership to implement AI. That number comes from a survey of 321 leaders conducted in October 2025 and published this week.
Second: Gartner separately predicts that by 2027, half of the companies that cut customer service staff because of AI will rehire people to perform similar functions -- often under different job titles.
Read those together: almost everyone is being pushed to replace humans with AI, and half of them will reverse course within two years. That is not a technology failure. It is a strategy failure. And small businesses have an opportunity to skip the expensive mistake entirely.
The Pressure Is Real and Growing
The Gartner survey paints a clear picture of where customer service leaders are focused in 2026. Their top priorities are improving customer satisfaction, operational efficiency, and self-service success rates. All three are areas where AI can genuinely help.
But the survey also reveals how the workforce is being reshaped. Nearly 80% of organizations plan to transition at least some agents into new roles as routine tasks get automated. And 84% of leaders plan to adjust hiring profiles and add new skills to the agent role.
The fastest-growing new function? Knowledge management. Fifty-eight percent of service leaders want to retrain agents as specialists who keep AI systems fed with accurate, updated content. That shift alone tells you something important: AI customer service tools are only as good as the information they are trained on, and someone human needs to maintain that information.
Why AI-First Customer Service Keeps Failing
The rehiring prediction is not theoretical. We have already watched the pattern play out.
Klarna made headlines when it replaced large portions of its customer service staff with AI. It then quietly reversed course and resumed hiring human agents after discovering the limitations firsthand. Air Canada was held legally liable when its chatbot gave a passenger incorrect information about bereavement fares. Cursor, the AI code editor, had its support chatbot fabricate company policies that did not exist.
These are not edge cases. They are the predictable result of deploying AI to cut headcount rather than to solve customer problems.
According to Intercom's 2026 Customer Service Transformation Report, 82% of senior leaders invested in AI for customer service in the past 12 months, and 87% plan additional investments in 2026. But as Kathy Ross, senior director analyst in Gartner's Customer Service and Support practice, noted: "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."
Meanwhile, research from Puzzel found that while 39% of companies resolve issues faster with AI and 83% of leaders say their self-service tools handle problems without a human agent, customers still report feeling unheard. Speed without empathy is not service.
What Small Businesses Should Actually Do
This is where being small becomes an advantage. Enterprise companies are locked into massive AI deployments they cannot easily reverse. A small business can be surgical about where AI helps and where humans stay in the loop.
Use AI for the boring stuff, not the hard stuff. Route tickets, auto-tag inquiries, draft initial responses, surface relevant knowledge base articles. These are the tasks where AI delivers immediate value without customer-facing risk. Let humans handle anything emotionally complex, ambiguous, or high-stakes.
Build your knowledge base before you deploy AI. The Gartner data on knowledge management is the most actionable finding in the entire survey. If your FAQ page is outdated, your return policy lives in someone's head, or your product documentation has gaps, fix that first. AI chatbots trained on bad information will confidently give wrong answers.
Measure what matters to customers, not just what matters to your budget. First-contact resolution rate. Customer effort score. Repeat contact rate. These tell you whether AI is actually improving the experience or just deflecting people into frustration loops. If you are only tracking cost-per-ticket, you are optimizing for the wrong thing.
Keep the human handoff seamless. The worst AI customer service experiences happen when customers fight through three layers of automation only to reach a human agent who has no context about the conversation. If you deploy any AI support tool, make sure it passes the full interaction history to the human agent when escalation happens.
The Augmentation Model Wins
Kim Hedlin, director of research in Gartner's Customer Service and Support practice, framed it well: "Leaders are not just deploying AI -- they are redesigning service models to ensure that technology enhances the customer experience while humans provide context, empathy, and judgment."
This is exactly the approach we advocate for small businesses integrating AI. The companies that will come out ahead are not the ones that replaced the most humans. They are the ones that made their humans more effective by removing tedious work and giving them better information at the moment they need it.
The pattern is consistent across every AI deployment we have seen succeed. Whether it is automating telephony, streamlining workflow orchestration, or building smarter self-service -- the winners treat AI as a tool for their team, not a replacement for it.
The Bottom Line
The next 18 months will see a wave of companies scrambling to rehire the customer service expertise they cut. That is an expensive lesson in confusing automation with strategy.
Small businesses do not have to learn it the hard way. Start with your knowledge base. Automate the repetitive tasks. Keep humans where empathy and judgment matter. And measure customer outcomes, not just cost savings.
Need help building an AI customer service strategy that actually works? Reach out to us. We help small businesses deploy AI in ways that make their teams stronger -- not smaller.