 Big businesses love AI, especially the AI powered chatbots. They love those for the same reason that they love any other automation tool because those tools have the potential to increase their profits massively. Because instead of them paying someone in the United States minimum wage to chat with customers online who have some questions or concerns about your product or service, they could just pay chat GBT or something similar to that. Pennies on the dollar compared to what that person's wage would be for API access and a few more pennies on the dollar for the cost of electricity to run that bot. In fact, this customer service model is actually even cheaper than just outsourcing the job to a call center in India. That's right. Even customer service positions that are outsourced to India are being made obsolete by AI chatbots. These chatbots don't complain. They don't go on strike. They don't file harassment lawsuits, ask for days off, get sick, show up late or get fed up one day and just decide to quit. But the thing about these chatbots is they aren't actually sentient. They don't have any consciousness. They aren't a real person. They just try to emulate a real person, which is obviously really annoying to customers that have to interact with them. A lot of clients in the U.S., they already get really annoyed by real people who just have foreign accents that work in customer service jobs because they feel like those non-native English speakers are not able to properly understand their problem. But the AI is literally unable to understand your problem. Sure, it has perfect spelling and perfect grammar, but that's probably pretty low on your list of priorities whenever you're contacting someone from customer service. However, there is a way that you, the customer, can take advantage of these AI chatbots to get huge savings or at the very least, you can have a little bit of fun because that's the most you can ask for some days in our ever-growing cyberpunk dystopia. So the most wholesome example that I've seen so far with this was probably Air Canada being forced to give a partial refund on a customer's plane tickets that they bought to travel for a funeral. Now, it's not just wholesome because of that specific circumstance with bereavement travel, but you know, if you've ever had to deal with airlines, right, especially airlines here in America, you know how difficult it is to get any kind of discount or to get any kind of good customer service experience out of them because at least here in the US, airlines actually have monopolies over certain areas, right? Like if you want to fly to some places in the country or basically forced to use one specific airline and they know that, which is why a lot of the time they try to end up screwing you. But sometimes airlines do offer discounts for bereavement travel. The thing about it just like everything else is that they usually make you jump through some hoops. There's some terms and conditions to how you can qualify for this bereavement travel and at the end of it, if you've jumped through all the hoops, if you've met all the terms and conditions, they might just write out deny you for that discount because again, you don't have that much choice when it comes to airlines. So what happened with this case with Air Canada is a customer needed to purchase a flight from Vancouver to Toronto to attend their grandmother's funeral. So when the customer visited the site, they were greeted by a chatbot and they asked the chatbot if Air Canada offered bereavement discounts and the bot said they did and it actually provided them with a link to the page for Air Canada's bereavement fares. Now, if you're in a chat with customer service, especially if it was a human customer service rep on the other side, you would probably want more than just a link, right? Because the customer could have gotten this same response even quicker by just googling the same exact question that they input into the chat window here. You know, you got to add that little personal touch. That's what customer service is all about, which is why the bot probably tried to explain the policy in its own words by saying if you need to travel immediately or have already traveled and would like to submit your ticket for a reduced bereavement rate, kindly do so within 90 days of the date your ticket was issued by completing our ticket refund application form. So, of course, this kind of response here makes it sound like you've got 90 days from the date that you purchased your plane tickets to get a partial refund for that reduced price. And I guess if you wanted to learn more about Air Canada's bereavement policy or other situations where they might give you a discount, then you can go ahead and click that link, right? But you probably don't want to click that link at this point because you just had it explained by customer service what the policy is. But you see, if the customer had clicked this link, they would have seen that the actual Air Canada bereavement policy is 10 days, right? Within 10 days, you have to request your refund or your request for a reduced amount. Otherwise, they're not going to help you out. So, when the customer was emailing Air Canada about this couple of months later, they thought they were well within the policy that the chatbot told them, but the airline tried to deny them. So, the customer took them to small claims court in Canada, I have to remind you, and it was decided that the airline was at fault and the customer was awarded $650.88 in damages. So, that's a nice little victory over the AI right there. Now, late last year, there was another case which unfortunately didn't produce any customer discounts. I mean, it probably would have been pretty awesome if it did, but it produced some lulls. The Chevy dealership with a GPT-4 powered chatbot. Now, part of what made this so ridiculous was how open the chatbot was. Like at first, the chatbot wasn't really just designed to talk about trucks. It was basically just open access to GPT-4. So, yeah, this was sort of a way for you to use GPT without actually having to pay for and sign up for a GPT account yourself. You could just hit up a random Chevy dealership and ask it to write some Python scripts for you or ask it to do some linear algebra or ask it if it has any strong opinion about trans women. But best of all, you could get it to sell you a 2024 Chevy Tahoe for $1. And that's a legally binding offer. No takesies, backsies. Now, of course, this didn't actually get the customer the 2024 Tahoe for a dollar, but this might actually start a trend where people are trying to hack customer-facing AI in order to get discounts from different businesses. I suspect this is going to be most common in fast food places or stores like Walmart, where some people are already exploiting flaws in the automated checkout kiosks to like get free burgers or free groceries or whatever. So what do you think? Will AI chatbots end up eroding customer trust with the companies that use them in place of employees? Will the losses that result from shrink or judgments against these companies be greater than the cost of just paying a human employee? Let me know in the comments below. And if you want to receive discounts online without exploiting an AI chatbot, check out my online store, base.win, where you can automatically save 10% at checkout when paying in Monero XMR. Have a great rest of your day.