Share with your CMO
Customers are routing around corporate AI investments and going straight to ChatGPT, and Gartner’s numbers make the scale of the problem hard to dismiss. A survey of 3,566 B2B and B2C customers found they are roughly three times more likely to use a third-party AI tool than a company-provided chatbot for service issues. Third-party GenAI use nearly doubled in the past year. Company chatbot use has been statistically flat since 2022. Service and support leaders are spending a median 12% of their 2025 budget on AI, the highest share of any business function surveyed, and only 24% report positive financial returns.
What this means for your business
The flatline on chatbot adoption after four years of investment is the telling detail here. If your customer experience org has been treating AI as a deflection channel, a way to answer questions cheaply before a human picks up, customers have already rendered their verdict. The 58% of GenAI users who’ve used AI to complete a task on their behalf, climbing to 74% in B2B, signals that the product customers want is an AI that acts, not one that talks. Companies whose chatbots can only retrieve answers are losing the comparison every time a customer opens a browser tab.
The structural problem is that most enterprise chatbots were designed around a cost-reduction frame rather than a capability frame. Deflect the call, save the agent-minute, declare success. That logic made sense when AI’s actual task completion ability was limited. It doesn’t hold now that a consumer can ask ChatGPT to draft a dispute letter, summarize a contract clause, or walk through a return process step by step. Gartner, whose advisory revenue depends on service leaders staying worried enough to buy research, has an obvious incentive to push the “redesign your journey” message rather than the blunter “you bought the wrong thing,” but the underlying data doesn’t require spin. The gap between third-party adoption and corporate chatbot adoption is a customer preference signal, not a marketing problem.
The decision this reframes isn’t whether to invest more in chatbots. It’s whether the chatbot budget should be redirected toward AI that sits inside workflows, booking engines, account portals, and document systems, places where action is possible, rather than a conversational layer bolted in front of a static knowledge base. CMOs who can make that case to their CIO counterpart now, before the 2026 budget cycle locks, are in a better position than those defending last year’s chatbot roadmap. I’d revise this view if company-provided chatbot adoption actually ticks up once more enterprises deploy genuinely agentic tools rather than FAQ wrappers, but that distinction hasn’t shown up in the adoption curve yet.
Concept deep-dive: Agentic AI
Agentic AI refers to systems that can take multi-step actions on a user’s behalf, completing a task rather than just answering a question, think booking a flight versus explaining how to book one. It matters here because the customer expectation gap Gartner is measuring is essentially the gap between conversational AI and agentic AI. Companies built chatbots. Customers are now comparing them to tools that act. Closing that gap requires back-end integrations, not better conversation design.
Based on reporting from Most Customers Don’t Care About Your AI Chatbot, originally published 2026-07-08 12:23:00.
