6 ms·

Anyone that has ever interacted with a customer service chat bot knows this is pretty stupid. I’m not sure why people (esp. those in the tech community) think

• tempestn 3y ago
  I don't know if that's going to be the case for long. GPT-4 is far more useful than almost any customer support rep I've dealt with.
• mupuff1234 3y ago
  It doesn't have to replace all people, just good enough to replace the more common cases.
  • lucb1e 3y ago
    Or frustrate people enough so they go and figure out the solution by themselves, or not file that complaint, or whatever it was they wanted to do but did not actually necessarily need assistance with, while giving the semblance of trying to be helpful Not saying this is a good thing, but then I'm not an upper manager with dollar symbols in their eyes
  • bobthepanda 3y ago
    In practice, places will ignore that it doesn't work for everything, cut off any sort of physical call centers, and then people just end up fed up with their services. I absolutely hate places aggressively adopting chatbots and phone bots.
    • mupuff1234 3y ago
      Perhaps, but that's also true with physical call centers, some of them are helpful and some are hell. Hopefully the costumers will rewards companies with better customer service. A good company will provide a good chat bot that will make it immediately clear when it can help and when it can't, with a clear and simple way to contact a human if needed.
      • ghaff 3y ago
        >Hopefully the costumers will rewards companies with better customer service. That assumes people will recognize those companies and pay a premium for their services.
        • neltnerb 3y ago
          If a company has really good products that never have issues, who would ever know whether they have good customer service? I kid, sort of, but I've seen awfully expensive equipment in use for years before anyone really stress tests the vendor support, and so much of reputation is just advertisements and luck. And the quality of support seems to change pretty quickly... though usually for the worse...
  • m463 3y ago
    I have seen that businesses over the last couple of years HAVE replaced some or all people. With amazon, I used to be able to email someone. Now I cannot. I have been forced to use a chat session. The chat session is partially automated. It could be completely automated, and I don't know if people will always be able to tell the difference. Tesla basically forces all service calls to the tesla mobile app. Getting a hotel is similar. web pages and phone calls are a minefield of dark patterns, where the customer is worn down and worn out by automated systems. There are exceptions, but it is a supreme battle of will and patience.
• smeej 3y ago
  Anyone who has ever interacted with a tier 1 customer service agent in an outsourced support farm knows the difference between that and a chat bot is negligible. True technical support can't be replaced yet, but thousands upon thousands of tier 1 human drones can be.
  • ke88y 3y ago
    > True technical support can't be replaced yet, but thousands upon thousands of tier 1 human drones can be. I disagree. Those agents are useless by design. They exist to do nothing in most cases, but none-the-less need to exist because of the regulatory + political + market backlash if they didn't. Chatbots don't fulfill the core requirement of being able to tell regulators and juries that you did, in fact, have an employee look into the customer's issue.
    • esafak 3y ago
      How do you know they're not just incompetent?
      • ke88y 3y ago
        Tomato, Tomato
      • bpm140 3y ago
        One person is incompetence. An entire staff is strategy.
  • avodonosov 3y ago
    I haven't, but saw this in movies: https://youtu.be/BqnFnWSYzaQ https://youtu.be/BqnFnWSYzaQ
  • johnnyworker 3y ago
    I once worked in a call center as tech support for an ISP that also did phone and TV. One day a women called about her continuing problems with the phone service crapping out. She had been in contact with us for months, and at some point she said she got the feeling people didn't take her seriously because she was a woman. She was a widow who was taking care of her 40 year old heavily disabled son, where any second there could be a life-threatening emergency requiring her to call an ambulance. I took down all the details, and, briefly detailed that she tried everything she could on her end. All previous tickets were closed with "seems fine", basically. Which ticked me off, but I tried to be somewhat professional about it, other than spending more time on that than my employer would have liked. But then I also did what was totally verboten, I wrote down her customer number, and a week later I checked in on her account again. As I expected, the ticket got closed again. So I did the SUPER MEGA verboten thing (because we got paid for inbound calls), I called her, asked if the problem still persists, it did. So I went through all the details again and wrote a super detailed ticket including the medical situation, that she had replaced the phone and everything else she did. I knew better than to be full on snarky, but I also did not hold back. I was fuming. 1-2 weeks later I checked again, it was solved, I called her, she said a tech had shown up and been super nice and helpful. I don't recall what the issue was, but they fixed it. I was so proud, and I couldn't even tell anyone about it. The only reason I did that was that I figured if they fire me from this minimum wage job I can get just another minimum wage job -- but I would not be able to live with myself to just go by the script in such cases. And I had plenty discussions with team and project lead about my average handling time, you bet. But they never fired me -- because I did solve problems, and with things that were standard procedure I was quick, I ended up quitting. A bot cannot do that, be disobedient. Too few people can even do that.
• gnicholas 3y ago
  I've tried Amazon's chat function and it's put me in an infinite loop on several occasions. I don't know why they didn't have an agent available (it was business hours). I was surprised that a company with Amazon's resources doesn't have a better handle on this, until I remembered that this is the company that doesn't let you say the reason is "wrong size" when returning clothing items. They're not stupid — they're trying to make things difficult for you.
  • sahila 3y ago
    Amazon support is amazing, there’s an option for chat or a receiving a call and always pick call.
    • gnicholas 3y ago
      Amazon's support can be amazing. I remember years ago, when every support call was answered by a knowledgeable and helpful employee in WA. That was amazing! More recently, I was lied to repeatedly by agents who told me that they were going to do something, or had done something already. I had to contact support many times, and the issue was only fixed after I was able to escalate to the executive support team, who confirmed I had been lied to repeatedly, and who were able to fix the problem.
  • madog 3y ago
    They also do not have an option to report an item they sent as counterfeit/fake which is amusing given how prevalent it is with their mixed bins between 3rd party items and legitimate goods. The closest related option is "not as described" or something like that. Even if you contact support chat that is the only option available to the advisers too. So it's actually impossible to report something as counterfeit. I always imagine some C-level exec looking at the stats and congratulating themselves on having no counterfeit items reported. Well done Bob! We did it again.
  • michaelcampbell 3y ago
    > a company with Amazon's resources How do you think they got to be that big?
• AussieWog93 3y ago
  I'm assuming that you're looking at it from the point of view of someone who only contacts tech support when human intervention is actually required. And honestly, probably 90% of people are like this. In terms of who actually takes up the majority of the support reps' time, however, is the 10% of people who are bored, lonely, genuinely mentally handicapped, whingers, and people who get really angry about some bizarre inane detail. I'd honestly love it if companies weren't so allergic to offering good, paid support or just dropping troublesome customers so that the rest of us could get through quickly.
  • addisonl 3y ago
    > probably 90% of people are like this I think you’re overestimating the average person.
  • tmpz22 3y ago
    > is the 10% of people who are bored, *lonely* I heard this is especially prevalent for suicide hotlines - because of the strict requirements to take and stay on such calls they have to put up with a lot of sexual harassment and its more or less swept under the rug by supervisors.
• Der_Einzige 3y ago
  Nope, the same people would recognize that AI chatbots are the best thing coming for education in a century: https://en.wikipedia.org/wiki/Bloom%27s_2_sigma_problem https://en.wikipedia.org/wiki/Bloom%27s_2_sigma_problem
  • svieira 3y ago
    IFF they produce facts and only facts they would STILL fail at their goal, because education is the transmission of virtue and machines cannot be virtuous. As a lossy interface on top of the global library that is the internet LLMs do just fine.
• gumballindie 3y ago
  > esp. those in the tech community It’s a vocal minority of the tech community that shares that opinion. The silent majority knows it’s just fud marketing. Personally when i encounter a customer support chat bit i immediately assume low quality services and products. When i meet a software engineer blown away but chatgpt’s “coding skills” i assume that engineer is not particularly competent. There are valid use cases and the tech has obviously got potential but we are not there yet.
  • jiggawatts 3y ago
    Few people are "blown away" by the absolute skill level of ChatGPT. It's more that there exists an AI that can do this at all. This just wasn't a thing three years ago, and the first truly capable version (GPT 4) only become publicly accessible this year. Some things I've seen GPT 4 do is "as good as human", often better than average. For example, spotting bugs or reverse-engineering code and then writing doc comments for it. The real power is not in having an AI as smart as you, it's having something "good enough" that can be deployed cheaply at scale to do rote, menial tasks. Review code, check for vulnerabilities, fix spelling mistakes in comments, etc...
    • webmaven 3y ago
      "The marvel is not that the bear dances well, but that the bear dances at all." — Russian proverb
    • chmod775 3y ago
      You want to trust that thing to review your code? It will often give you eloquent explanations, trying to convince you that code has a certain bug, while completely missing the actual bug. You'll be gaslit into chasing down dead ends, distracted from actual issues. Example: "In JS, create a Promise that is resolved with some emitted value from an EventEmitter, or rejected if an 'error' event is emitted first. Write that concisely, using '.once' and only removing the other event respectively." Fairly everyday thing in node. Every competent programmer would grasp the intent here pretty quick and write correct code, but somehow it's generally beyond ChatGPT. Example ChatGPT output (comments mine): function eventToPromise(emitter, event) { return new Promise((resolve, reject) => { emitter.once('error', reject); // memory leak: not removing the other listener emitter.once(event, (data) => { emitter.removeListener('error', reject); resolve(data); }); }); } Come on now... I just start the whole prompt over to let it try again: function promiseFromEvent(emitter, eventName) { return new Promise((resolve, reject) => { const resolveHandler = (data) => { emitter.removeListener(eventName, errorHandler); // memory leak: should be removeListener('error', errorHandler) resolve(data); }; const errorHandler = (error) => { emitter.removeListener(eventName, resolveHandler); reject(error); }; emitter.once(eventName, resolveHandler); emitter.once('error', errorHandler); }); } Now I prompt "You have a bug in there, fix it." ChatGPT: "Apologies for the oversight. You are right; there is a bug in the previous code. The issue is that if the 'error' event is emitted first, both the 'error' and 'data' event listeners should be removed, and the Promise should be rejected." Nope. The error event listener is actually the only listener working correctly... It proceeds to introduce a cleanup function that removes both listeners (despite keeping '.once'), fixing the bug by accident, not intent. Asking it to change it so that the code adheres to the original prompts starts a downward spiral.
• bpm140 3y ago
  Every time I am forced to endure shitty hold music for an extended period of time, my thought is, “let the CEO sit on hold for ten minutes and see how fast this gets changed.” What this means is I still want to believe that crap customer experience is due to a disconnected CEO, when we all know it’s a feature, not a bug.
• onetokeoverthe 3y ago
  [dead]