A majority (64%) of 5,728 people surveyed by Gartner in December 2023 said they would prefer companies not to use AI in their customer service. Additionally, 53%...
If it worked for most shit and escalated to a human when it actually needed to, reliably, I'd be fine with it.
I don't believe there's a realistic chance that there's a lot of overlap between the people willing to invest to actually do it properly and the people paying for AI instead of people though.
I extremely hate this idea. I I already hate the automated systems that are definitely designed to make you give up just trying to talk to an actual human being. Hopefully, we can get more lawsuits around the world like the Air canada one where they are liable for any bs the ai decides to make up, along with actual laws saying the same. Hopefully, it would discourage them.
The point of modern "customer service" is to NOT provide customer service. If you can drag out the conversation to the point where the caller rage-quits in frustration, then the company can avoid spending any money on fixing any problems they've caused.
I had the displeasure of being called by one from a vendor. It pissed me off that they couldn't be bothered to pick up the phone and call using a human, with how much we paid them. I canceled that contract and went with a different vendor, and let the sales team know exactly why. LLMs have their place, but my time is not the waste bin.
Already out there in certain ways. There's a restaraunt near me that uses an automated system to collect orders in the drive-thru, and puts them into the system incorrectly.
At least that's what seems to be its purpose, because it does that really well. That, and piss people off.
Storytime! Earlier this year, I had an Amazon package stolen. We had reason to be suspicious, so we immediately contacted the landlord and within six hours we had video footage of a woman biking up to the building, taking our packages, and hurriedly leaving.
So of course, I go to Amazon and try to report my package as stolen.... which traps me for a whole hour in a loop with Amazon's "chat support" AI, repeatedly insisting that I wait 48 hours "in case my package shows up". I cannot explain to this thing clearly enough that, no, it's not showing up, I literally have video evidence of it being stolen that I'm willing to send you. It literally cuts off the conversation once it gives its final "solution" and I have to restart the convo over and over.
Takes me hours to wrench a damn phone number out of the thing, and a human being actually understands me and sends me a refund within 5 minutes.
Companies don’t want to provide actual service for problems. That costs money. They want you to give up.
Customers hate anything that actually gets between them and someone that can actually help. Not shitty, complicated automated phone menus. Not some underpaid stooge who refuses to da anything except read from a mandatory customer service script. And not AI, which will combine both of the worst aspects of automation and scripted service along with a cheerful idiot that will spare no effort to direct you away from the nearest actual assistance.
There's this boomer obsession with making it listen to human speech...
Nobody under 40 wants to use human speech to talk to an AI. We don't want to us human speech to talk to humans most of the time, especially if we don't know them.
But they always want to jam an AI into areas where human speech is the main communication method.
The absolute last place AI should have been deployed is answering a phone call. Because that is the last resort for most people, but the boomers calling the shots still think that's people's go to move before trying anything else
I do like it in the sense that people HATE working in customer service. Because people have zero respect and customers make your job day miserable all the time.
Is one of the places where people deserve getting a hallucinating robot as a vengeance for how bad they treated people that worked there.
Automated phone systems have been a thing for decades. They are notoriously shitty and adding a layer of “friendly AI” on top of that shitty system doesn’t bode well.
Realistically we only dislike it because it’s a half baked solution. I know that if those LLMs actually did anything useful we wouldn’t mind them. But all these LLMs do is spam the documentation, which is already on the vendor website anyway.
The other day I was able to take care of what I thought was a reasonably complicated customer service issue through an automated assistant.
I take a daily prescription medication and it’s on automatic refill. However now and then I forget to take my pill and then I have an extra. After years of this I found myself with 20-30 pills left when my next bottle was ready.
So I tried to call the pharmacy and say hey that bottle of pills you have waiting for me? I still need it, but not for about 3 weeks. Can you push my entire schedule back that much but otherwise keep the pace the same?
Turned out I was able to do this just by listening to menus, selecting from multiple choice, and entering numbers for dates.
I was so satisfied! I don’t want to talk to a human if I can possibly help it. I’d much rather deal with an automated system as long as it can do what’s needed. The problem is that most of them can’t. But then again most customer service humans are useless too, so…
There’s a NYT article somewhere, and I’ve been desperately trying to find it, about a woman who worked as some kind of real estate(?) call center AI augmenter. Essentially people would call in about listings or something, and she had to step in when the AI went off the tracks or didn’t know how to answer questions, matching its tone/inflection while refusing to acknowledge that there was a human stepping in. She ended up being super burnt out from the job. So the whole system was just super redundant, awful for the people working there, and as we’ve come to expect from AI, just a half-baked turd sold to some MBAs for a mint.
I'm already pissed with bots, had to call my ISP yesterday because my internet was spotty, I couldn't talk to a single human, the bot was walking me through the tired modem restart, and then it ended the call and asked for me rate it even though it didn't solve anything!
Unless they hate it enough to ditch a business or service in great enough numbers that it costs the business more money than they save by outsourcing to a computer, people had better get used to it.
Yeah it turns out that using a statistical model to handle customer service leads to a degraded customer experience, because statistical models aren't humans and lack many human attributes.
The telephone services i had to use in my Lifetime were so insanely bad, thats one of the few things llm could do better than an underpaid person who has no will to live anymore and got yelled at for hours. This shit needs to end.
Unpopular opinion and rant:
Us, the consumers, brought this on ourselves.
Not intentionally but it was a slippery slope.
No one I know did ever ask the sales representative "does your customer support answer within 5 minutes and will I always reach a representative with att least 10 years of experience, that has the authority to make real decisions?". No, but we were all very interested about the pricing of the service/product.
Then these "Please press 1 for...." happened.... and no one of us really cared about the change because the service providers offered a much lower price than the ones with customer support representatives with 20 years of experience.
Since all of us went for the cheapest provider, the other ones had to cut cost to be able to offer their service on a competitive price level.
So then there were no one offering competent support with representatives that knew their shit.
And it slowly continued to go downhill...
So here we are with shitty services, which we pay for, where we all are treated as cattle.
If people at least started to ask for better customer support there would someone, who wants to climb the corporate ladder, creating a PowerPoint presentation with a real VIP Service Level.
Of course it will cost more money, because real people cost money, but we would att least get what we want.
But no. Consumers will still go for the lowest price.
I love the idea of useful and improved AI-automations, based on the fact every site currently has a “customer service robot” that have never once helped with a resolution at all.
IMO it can eliminate a huge amount of support queries and leave the important stuff to the actual agent if done right. …with the caveat of yeah fuck AI if it’s fully instead of agents.
While I’d prefer to just speak to a human, I’d much prefer AI over the status quo of dead dumb automated systems that just keep looping through the same preset options until you get enraged and give up or mash zero
AI can only give you the options it was programmed to give. A Human is able to actually think and find a solution or direct you to a solution.
Your options are less with AI for customer service.
AI works best for applications that it is tailored for. But expecting it to "think" like Humans do is so far off. AI is being fed so much biased information and that is not "thinking" or learning.
It is now at the point where we need to ask how they plan on handling complaints and problems. And if the answer is not correct, go somewhere else. Up till now this was never something we needed to worry about
I wonder if it would actually materialise, consisting the recent case where an airline company's AI chatbot promised a refund that didn't exist, but were expected to uphold that promise.
That risk of the bot offering something to the customer when the company would rather they not, might be too much.
It seems more likely that companies will either have someone monitoring it, and ready to cut the bot off if it goes against policy, or they'll just use a generated voice for a text interface that the client writes into, so they don't have that risk, and can pack more customers per agent at a time in.
Fair point but no one wants to deal with my incessant whining, and you couldn't pay me enough to deal with it. Sometimes a scratching post is what the cat needs.
The only good thing about is that, with most companies, if you need a refund for something under $10, it generally just gives you the refund and sets you off on your merry way.
Honestly, I've used some pretty decent AI chatbots. They can help you with basic questions and contact you with a human for things that require it or if you ask for it. Chatbots that don't let you talk to a human on the other hand, those are awful.
An exceptionally well trained AI customer service has the potential to be amazing.
I only call or try to chat/email with customer service if something has gone way wrong - like outside the typical customer service capability of assistance.
If an AI can realize that my problem is human worthy and escalate it faster, that would save me time in the chat queue talking with someone who barely knows my native language.
Alas, AIs will be poorly trained, so the bad-english CS reps will still be right behind the AI interface waiting for me.
Let's be honest here: they want a human to abuse. They want to be shitty to and verbally assault someone that they view as being "lower" than them. If the AI works well (a different conversation) then people will get over any trepidation they have rather quickly. The people that are legitimately upset will just miss having someone to put down for "only" working customer service.
I would like it if it actually did something for me, (like automatically doing x, y, or z to my account on the backend based on my request) but instead it just feels like every one of the "AI-Powered" support bots is designed to try and make it as hard as possible for me to actually get anything done.
Anyone who ever tried to solve any problem and gets stupid responses on ai chat instead of making it easy to reach a real person that can solve it in seconds knows the pain.