From the article: Customer care has become one of the most notorious business failures of the digital age, and everyone knows it. Now, artificial intelligence threatens to take this horror show of impersonal, unreliable service to a whole new level.
Within a couple of years, 50 percent of the organizations that had planned to replace their customer service personnel with AI models are expected to reverse their decision. According to a recent survey from Gartner, the original goals were overly ambitious – and ultimately unachievable. The transition to an AI-focused business world is proving to be far more challenging than initially anticipated.
In March 2025, the US research firm surveyed 163 leaders in the customer service and support industry. Nearly all respondents (95 percent) now say they plan to retain human workers while “strategically” evaluating what role AI technologies can realistically play within their organizations.
Kathy Ross, senior director analyst at Gartner, noted that while AI has the potential to transform customer service, it is not a miracle solution. Human interaction is still essential in many situations, especially when customers reach the end of a frustrating experience and need real help with a newly purchased product that isn’t working as expected.
Festering-Fecal on
When McDonald’s couldn’t make ai work that should have been a red flag for everyone
Roccet_MS on
Shocking, but I’m sure those wise and wealthy CEOs never had to deal with customer service via phone. Therefore they didn’t know that people dislike talking to AI even more compared to incompetent people on the phone.
snowbyrd238 on
They’re using the technology backwards. You don’t use AI to replace the people, you use it to replace the support structure.
Baconpwn2 on
Have these CEOs ever played with AI agents before? It takes time and effort to get the results you want. What client is ever going to be patient enough to deal with that?
TraditionalBackspace on
They will do it as soon as it saves them money. Guaranteed.
AstroZeneca on
For now.
The current conversation around AI is ridiculous. “AI can do everything, it will put people out of work” or “AI can’t do anything, it’s over-hyped and poses no risk”.
They’re both wrong, it’s a promising technology that, in many cases, has been adopted too early. Judging AI by what it can do in 2025 is like predicting a person’s future based on how they behave as an infant. The conversation will be different in 5 years, and in 10 years.
phobox91 on
And not for moral or social purposes, Just because they are not seeing the profit they were expecting.
talrich on
The sad thing is that automated voice recognition systems are so bad and frustrating that a huge number of customers are understandably angry by the time they reach an actual human.
Harbinger2001 on
I work in software development and there is still a lot that has to be figured out before we can reliably use AI for automating customer interactions and automation. Using an unpredictable algorithm in the core of your automation is insanity and goes against why computers became so important in the first place. Repeatable processes at massive scale is not achievable with LLMs and things like AI Agents. It can be deployed in controlled environments for business internal use but putting it customer facing is courting disaster.
Presently_Absent on
Every chatbot or AI I’ve ever encountered as a support replacement has been frustrating and put me in a shitty mood for when I finally talk to a person, because they are basically useless.
Jnorean on
It had to happen. Companies listen to the AI hype and believe it. This creates unrealistic expectations of what the AI can do. They create a system utilizing its weaknesses and ignoring its strengths. When they attempt to roll out the AI they are baffled by what it really does and stop it. Now they won’t go back to AI technology believing it is all nonsense. This should stop the mantra of 60% of workers being replaced by AI next year so often Hyped by the AI creators. Good for the workers and not so good for the AI companies.
thelentil on
We’re going to have maybe 5 years of it being forced on us on all fronts by market trends then I hope people will be hungry enough for human voices, contact and craft/art at that point that we all revert to being luddites. A super optimized AI-based world sounds like an absolute fucking nightmare.
generalmandrake on
I own a business and someone tried to sell me on an AI that would interact with clients, answer questions and schedule things until my calendar. I said absolutely no way. AI can do amazing things but it also acts like a partially brain damaged person and I would never put it into a position where it could do real harm to my business and I do not trust the people programming these things to actually ensure that it wouldn’t do real harm.
14 Comments
From the article: Customer care has become one of the most notorious business failures of the digital age, and everyone knows it. Now, artificial intelligence threatens to take this horror show of impersonal, unreliable service to a whole new level.
Within a couple of years, 50 percent of the organizations that had planned to replace their customer service personnel with AI models are expected to reverse their decision. According to a recent survey from Gartner, the original goals were overly ambitious – and ultimately unachievable. The transition to an AI-focused business world is proving to be far more challenging than initially anticipated.
In March 2025, the US research firm surveyed 163 leaders in the customer service and support industry. Nearly all respondents (95 percent) now say they plan to retain human workers while “strategically” evaluating what role AI technologies can realistically play within their organizations.
Kathy Ross, senior director analyst at Gartner, noted that while AI has the potential to transform customer service, it is not a miracle solution. Human interaction is still essential in many situations, especially when customers reach the end of a frustrating experience and need real help with a newly purchased product that isn’t working as expected.
When McDonald’s couldn’t make ai work that should have been a red flag for everyone
Shocking, but I’m sure those wise and wealthy CEOs never had to deal with customer service via phone. Therefore they didn’t know that people dislike talking to AI even more compared to incompetent people on the phone.
They’re using the technology backwards. You don’t use AI to replace the people, you use it to replace the support structure.
Have these CEOs ever played with AI agents before? It takes time and effort to get the results you want. What client is ever going to be patient enough to deal with that?
They will do it as soon as it saves them money. Guaranteed.
For now.
The current conversation around AI is ridiculous. “AI can do everything, it will put people out of work” or “AI can’t do anything, it’s over-hyped and poses no risk”.
They’re both wrong, it’s a promising technology that, in many cases, has been adopted too early. Judging AI by what it can do in 2025 is like predicting a person’s future based on how they behave as an infant. The conversation will be different in 5 years, and in 10 years.
And not for moral or social purposes, Just because they are not seeing the profit they were expecting.
The sad thing is that automated voice recognition systems are so bad and frustrating that a huge number of customers are understandably angry by the time they reach an actual human.
I work in software development and there is still a lot that has to be figured out before we can reliably use AI for automating customer interactions and automation. Using an unpredictable algorithm in the core of your automation is insanity and goes against why computers became so important in the first place. Repeatable processes at massive scale is not achievable with LLMs and things like AI Agents. It can be deployed in controlled environments for business internal use but putting it customer facing is courting disaster.
Every chatbot or AI I’ve ever encountered as a support replacement has been frustrating and put me in a shitty mood for when I finally talk to a person, because they are basically useless.
It had to happen. Companies listen to the AI hype and believe it. This creates unrealistic expectations of what the AI can do. They create a system utilizing its weaknesses and ignoring its strengths. When they attempt to roll out the AI they are baffled by what it really does and stop it. Now they won’t go back to AI technology believing it is all nonsense. This should stop the mantra of 60% of workers being replaced by AI next year so often Hyped by the AI creators. Good for the workers and not so good for the AI companies.
We’re going to have maybe 5 years of it being forced on us on all fronts by market trends then I hope people will be hungry enough for human voices, contact and craft/art at that point that we all revert to being luddites. A super optimized AI-based world sounds like an absolute fucking nightmare.
I own a business and someone tried to sell me on an AI that would interact with clients, answer questions and schedule things until my calendar. I said absolutely no way. AI can do amazing things but it also acts like a partially brain damaged person and I would never put it into a position where it could do real harm to my business and I do not trust the people programming these things to actually ensure that it wouldn’t do real harm.