New ReverseLookup data suggests customers are not rejecting artificial intelligence in customer service. They are rejecting support that feels repetitive, unhelpful and incapable of solving real problems.
The promise of AI customer support was simple: faster answers, shorter queues and instant help at any hour. Instead, many customers say they are getting something else entirely — replies that arrive quickly but fail to move the conversation forward.
The problem is not that the software is artificial. It is that too often it feels bad at customer service.
A new ReverseLookup survey of 7,567 respondents across Europe, the U.S. and Latin America suggests that frustration with AI support is driven less by automation itself than by the quality of the experience. 80% of respondents said they find AI customer support frustrating when trying to solve a serious or time-sensitive problem. 67% said automated systems often give answers that feel generic or unrelated to the issue. 63% said they had been asked to rephrase a problem that already seemed clear.
Those numbers point to a competence problem before they point to a technology problem.
Customers rarely expect a chatbot to solve every situation. They do expect it to understand what they have already explained. When a system responds confidently without demonstrating that it has understood the problem, speed stops feeling like an advantage. It feels like wasted time.
The pattern is familiar. Someone reports a locked account and is asked whether they want to reset a password. A traveler trying to change a booking is sent back to the same policy page they have already read. A customer reporting suspected fraud is asked to choose from a list of categories that do not match the situation. The responses are fast, polite and often completely unhelpful.
ReverseLookup found that 68% of respondents had been stuck with a chatbot or automated phone system that could not solve their problem. 58% said AI support repeated information they had already provided instead of moving the case forward. 52% said chatbot replies made them feel the company was not taking their issue seriously.
Customers are not evaluating AI models. They are evaluating customer service. And many increasingly describe today's support bots as repetitive, forgetful and incapable of handling anything beyond a scripted request.That frustration becomes even stronger when the chatbot is not just ineffective but unavoidable.
ReverseLookup found that 61% of respondents said it was difficult to reach a human operator even after the automated system failed. 57% said they had abandoned a complaint, request or support conversation because the process became too frustrating.
At that point, the problem is no longer simply that the AI performs poorly. The problem is that the poorly performing AI has become the gatekeeper.This helps explain one of the survey's most striking findings. 72% of respondents said they would rather wait longer for a human agent than receive an instant AI response that cannot actually solve the issue.
That figure is often interpreted as a rejection of automation. It is better understood as a rejection of bad automation.
Customers are not choosing slower service because they enjoy waiting. They are choosing the option they believe is more likely to produce a real solution. After enough generic replies, repeated questions and failed conversations, a longer wait begins to feel more efficient than an instant answer.
The consequences extend beyond a single support interaction. ReverseLookup found that 46% of respondents said a bad automated support experience made them trust a company less.
That may be the cost businesses underestimate. AI support is frequently measured through shorter queues, lower staffing costs and fewer support tickets. Customers measure something different. They judge whether the company understood the problem, whether the response changed anything and whether someone took responsibility when automation failed.
None of this suggests AI has no place in customer service. For routine tasks, it can be extremely effective. It can answer straightforward questions, surface account information, summarize previous conversations and help human agents work more efficiently.
The problem begins when companies mistake fluent conversation for competent support. A response is only useful if it helps solve the problem. Customers increasingly recognize the difference.
ReverseLookup found that 64% of respondents said companies should be required to offer a clear human-support option for account access, fraud, travel, service outages and other high-stakes problems. That finding is not anti-AI. It reflects an expectation that automation should make customer service better, not simply cheaper.
The companies most likely to earn customer trust will not be those with the fastest chatbot. They will be the ones whose AI knows its limits, solves routine problems well and hands complex ones to a person before frustration turns into distrust.
Customers do not expect AI to be human. They expect it to be useful. Right now, many believe it is neither.
About ReverseLookup
ReverseLookup is a multi-input verification platform for phone numbers, emails, and images. Built for everyday use, ReverseLookup.com enables users to assess unfamiliar contacts, investigate questionable profiles, and identify potential fraud across key digital channels. It combines reverse search methods with open-source intelligence (OSINT) to offer a direct, accessible way to review digital identities and make informed decisions online.
Media Contact
ReverseLookup
Ashleigh Thomas
PR Manager
pr@reverselookup.com
The problem is not that the software is artificial. It is that too often it feels bad at customer service.
A new ReverseLookup survey of 7,567 respondents across Europe, the U.S. and Latin America suggests that frustration with AI support is driven less by automation itself than by the quality of the experience. 80% of respondents said they find AI customer support frustrating when trying to solve a serious or time-sensitive problem. 67% said automated systems often give answers that feel generic or unrelated to the issue. 63% said they had been asked to rephrase a problem that already seemed clear.
Those numbers point to a competence problem before they point to a technology problem.
Customers rarely expect a chatbot to solve every situation. They do expect it to understand what they have already explained. When a system responds confidently without demonstrating that it has understood the problem, speed stops feeling like an advantage. It feels like wasted time.
The pattern is familiar. Someone reports a locked account and is asked whether they want to reset a password. A traveler trying to change a booking is sent back to the same policy page they have already read. A customer reporting suspected fraud is asked to choose from a list of categories that do not match the situation. The responses are fast, polite and often completely unhelpful.
ReverseLookup found that 68% of respondents had been stuck with a chatbot or automated phone system that could not solve their problem. 58% said AI support repeated information they had already provided instead of moving the case forward. 52% said chatbot replies made them feel the company was not taking their issue seriously.
Customers are not evaluating AI models. They are evaluating customer service. And many increasingly describe today's support bots as repetitive, forgetful and incapable of handling anything beyond a scripted request.That frustration becomes even stronger when the chatbot is not just ineffective but unavoidable.
ReverseLookup found that 61% of respondents said it was difficult to reach a human operator even after the automated system failed. 57% said they had abandoned a complaint, request or support conversation because the process became too frustrating.
At that point, the problem is no longer simply that the AI performs poorly. The problem is that the poorly performing AI has become the gatekeeper.This helps explain one of the survey's most striking findings. 72% of respondents said they would rather wait longer for a human agent than receive an instant AI response that cannot actually solve the issue.
That figure is often interpreted as a rejection of automation. It is better understood as a rejection of bad automation.
Customers are not choosing slower service because they enjoy waiting. They are choosing the option they believe is more likely to produce a real solution. After enough generic replies, repeated questions and failed conversations, a longer wait begins to feel more efficient than an instant answer.
The consequences extend beyond a single support interaction. ReverseLookup found that 46% of respondents said a bad automated support experience made them trust a company less.
That may be the cost businesses underestimate. AI support is frequently measured through shorter queues, lower staffing costs and fewer support tickets. Customers measure something different. They judge whether the company understood the problem, whether the response changed anything and whether someone took responsibility when automation failed.
None of this suggests AI has no place in customer service. For routine tasks, it can be extremely effective. It can answer straightforward questions, surface account information, summarize previous conversations and help human agents work more efficiently.
The problem begins when companies mistake fluent conversation for competent support. A response is only useful if it helps solve the problem. Customers increasingly recognize the difference.
ReverseLookup found that 64% of respondents said companies should be required to offer a clear human-support option for account access, fraud, travel, service outages and other high-stakes problems. That finding is not anti-AI. It reflects an expectation that automation should make customer service better, not simply cheaper.
The companies most likely to earn customer trust will not be those with the fastest chatbot. They will be the ones whose AI knows its limits, solves routine problems well and hands complex ones to a person before frustration turns into distrust.
Customers do not expect AI to be human. They expect it to be useful. Right now, many believe it is neither.
About ReverseLookup
ReverseLookup is a multi-input verification platform for phone numbers, emails, and images. Built for everyday use, ReverseLookup.com enables users to assess unfamiliar contacts, investigate questionable profiles, and identify potential fraud across key digital channels. It combines reverse search methods with open-source intelligence (OSINT) to offer a direct, accessible way to review digital identities and make informed decisions online.
Media Contact
ReverseLookup
Ashleigh Thomas
PR Manager
pr@reverselookup.com