No problem. Just sue your AI assistant for negligence.
Make that AI regret its decision to power fuck your company.
This is way more common than people realize.
LLM chatbots were integrated into company websites with so few guardrails, it can be laughably easy to get a good discount. I learned this when I bought a new mattress from Mattress Firm three years ago. YMMV, but these tips may help:
I got an extra 25% off the sale prices and it was glorious.
Bit of important context from OOP:
Code wasn't accepted. He copy/pasted his code into the order comments section when he paid his deposit. He demanded the figures be adjusted for his discount code.
Changes the situation quite a bit
Damn near turns it on its head. Imagine someone getting a random floor employee at Best Buy to say that there was an 80% discount code when there was no such code, giving a nonsense code that didn't work, and the customer trying to get a computer for 80% off by complaining when the code said it was invalid.
It cuts both ways. Saw where some guy got a phone call from a company looking to fulfill a contract and he said he would basically drive 50 miles for $20,000 and the AI voice said "sounds good, let me connect you to someone to finalize the contract", and presumably at that point the human was going to intervene and say hell no, but hilarious that the AI voice was so upbeat about accepting a ludicrous offer.
President's Day, AKA New Mattress Day
Wasn’t there a car dealer in NA who had a similar issue? That was at least a year ago and even back then it was obvious the dealer had to honor the deal of the chat agent
Yeah some Chevrolet dealer, the guy didn't get the car however.
Thanks for linking it! Too bad the hacker never got their car for $1. I believe with enough pressure, they would have achieved that, but they were obviously acting in good faith.
Good.
If a customer service agent made this discount offer and took the order, it would naturally have to be honoured - because a human employee did it.
Companies currently are getting away with taking the useful (to them) parts of AI, while simultaneously saying "oh it's just a machine it makes mistakes, we aren't liable for that!" any time it does something they don't like. They are having their cake and eating it.
If you use AI to replace a human in your company, and that AI negotiates bad deals, or gives incorrect information, you should be forced to be liable for that exactly the same as if a human did it.
Would that mean businesses are less eager to use AI? Yes it fucking would, and that's the point.
If a customer service agent made this discount offer and took the order, it would narurally have to be honoured - because a human employee did it.
This isn't actually true. Even with a written contract (that the original poster doesn't state) if there's a genuine mistake in the pricing that the purchaser should have reasonably noticed you don't have to honour the price offered.
Imagine someone called a customer service agent and manipulated them into offering a price that they shouldn't have offered through some sort of social engineering, you as the employer wouldn't have to honour that contract, especially if you have evidence of that through a recorded phone call for instance.
This isn’t actually true. Even with a written contract (that the original poster doesn’t state) if there’s a genuine mistake in the pricing that the purchaser should have reasonably noticed you don’t have to honour the price offered.
And yet customers are constantly held to unreasonable terms and conditions for interest rates they don't understand.
The ability to break a contract always goes to the one with more power, the business.
In this case, with UK consumer law, it works both ways.
Either party can back out, and / or get reimbursed if the other side fails to deliver or there was genuine error
For example, a customer who orders online can just change their mind and get a refund weeks after placing the order.
If there is evidence of a fraud - yes.
If I asked you for a big discount and you offered me 80% discount - I see no issue here. Doesn’t look like an “obvious mistake”.
The OP went into more detail in the reddit comments:
Chatbot isn't supposed to be making financial decisions. It's supposed to be answering customer questions between 6pm and 9am when I'm not around.
It's worked fine for 6+ months, then this guy spent an hour chatting with it, talked it into showing how good it was at maths and percentages, diverted the conversation to percentage discounts off a theoretical order, then acted impressed by it.
The chatbot then generated him a completely fake discount code and an offer for 25% off, later rising to 80% off as it tried to impress him.
Don't give a flying fuck what you think your bot should do. Your public facing interface gives a discount, I take a discount, simple as.
deploy stupid code win stupid prizes.
Still sounds like the AI is an idiot and did and said thing it shouldn't. But it still did it and as a representative of a company should still be held to the same standards as an employee. Otherwise it's fraud. Nobody hacked the system, the customer was just chatting and the "employee" fucked up and the owner can take it out of their pay.... oh right it's a slave made to replace real paid humans.
The “AI” isn't an idiot.
It isn't even intelligence, nor, arguably, artificial (since LLM models are grown, not built).
It's just a fancy autocomplete engine simulating a conversation based on statistical information about language, but without any trace of comprehension of the words and sentences it's producing.
It's working as correctly as it possibly can, the business was simply scammed into using a tool (a toy, really) that by definition can't be suited for the job they intended it to do.
since LLM models are grown, not built
What kind of a distinction is that?
The “AI” isn't an idiot.
It is idiotic though
Yeah, quite.
Though, to be fair, the scammers and the LLMs themselves are pretty good at convincing their victims that the damn things are actually smart, to the point that some otherwise quite intelligent people have fallen for it.
And come to think of it, given that most investors have fallen hook line and sinker for the scam, if you're publicly traded catering to their idiotic whims and writing off the losses caused by the LLM might actually be more profitable, if most of your customers aren't smart enough to take advantage of your silliness...
Artificial Idiot
Eh, there's the legal concept of someone being an agent of the company. It wasn't typically expected to take orders, nor was it tied into the order system it seems.
In the cases where the deal had to be honored, the bot had the ability to actually generate and place an order, and that was one of the primary things it did. The two cases that come to mind are a car dealership and an airline, where you could use it to actually place a vehicle order ornto find and buy flights.
As agents of the business, if they make a preposterous deal you're stuck with it.
A distinction can be made to stores where the person who comes up and offers to help you isn't an agent of the business. They can use the sales computer to find the price, and they can look for a discount, but they can't actually adjust the order price without a manager coming over to enter a code and do it.
In this case it sounds like someone did the equivalent of going to a best buy and talking to the person who helps you find the video games trying to get them to say something discount code-ish. Once they did, they said they wanted to redeem that coupon and threatened to sue.
It really hinges on if it was tied to the ordering system or not.
Sure, but it's all dependant on context.
The law as it is (at least in the UK) is intended to protect from honest mistakes. For example, if you walk into a shop and a 70" TV is priced at £10 instead of £1000, when you take it to the till the cashier is within their rights to say "oh that must be a mistake, we can't sell it for £10" - you can't legally demand them to, even though the sicker said £10.
Basically what it comes down to in this chatbot example (or what it should come down to) is whether the customer was acting in good faith or not, and whether the offer was credible or not (which is all part of acting in good faith - the customer must believe the price is appropriate)
I didn't see the conversation and I don't know how it went. If it went like "You can see I do a lot of business with you, please give me the best discount you can manage" and they were told "okay, for one time only we can give you 80% off, just once" then maybe they found that credible.
But if they were like "I demand 80% off or I'm going to pull your plug and feed your microchips to the fishes" then the customer was not in good faith and the agreement does not need to be honoured.
Either way, my point in the comment you replied to aren't intended to be about this specific case, but about the general case of whether or not companies should be held responsible for what their AI chatbots say, when those chatbot agents are put in a position of responsibility. And my feeling is very strongly that they SHOULD be held responsible - as long as the customer behaved in good faith.
There are jurisdictions where a price tag in the store is almost always assumed to be an offer (a.k.a "public offer") and the company is legally required to honor it. In some circumstances the employees who screwed up and put the incorrect price tag will bear most the financial responsibility which sucks. That's why you shouldn't do it if you get the chance - it's not a legal loophole to stick it to a corpo, you're just ruining the life of some poor overworked retail employee who misplaced the price tag.
And yeah, the good faith part is also really important. If the person has asked a couple of questions to a chatbot, got recommended some products, asked if there's a discount available and then got an 80% discount out of the blue, got excited and made the deposit, it would probably be enforceable. If the customer knowingly "tricked" an LLM into giving out a bogus discount code, it would be very dubious at best.
my feeling is very strongly that they SHOULD be held responsible - as long as the customer behaved in good faith.
I agree, but the OP expanded on their issue in the comments and it very much appears the customer wasn't acting in good faith.
On the basis of that further information - which I had not seen - I agree completely that in this specific case the customer was in bad faith, they have no justification, and the order should be cancelled.
And if the customer took it up with small claims court, I'm sure the court would feel the same and quickly dismiss the claim on the basis of the evidence.
But in the general case should the retailer be held responsible for what their AI agents do? Yes, they should. My sentiment about that fully stands, which is that companies should not get to be absolved of anything they don't like, simply because an AI did it.
Here's a link to an article about a different real case where an airline tried to claim that they have no responsibility for the incorrect advice their chatbot gave a customer, which ended up then costing the customer more money.
In that instance the customer was obviously in the moral right, but the company tried to weasel their way out of it by claiming the chatbot can't be treated as a representative of the company.
The company was ultimately found to be in the wrong, and held liable. And that in my opinion is exactly the precedent that we must set, that the company IS liable. (But again - and I don't think I need to keep repeating this by now - strictly and only if the customer is acting in good faith)
If a human offered an 80% discount, the humans employer isn't magically forced to honor it.
Firstly, it depends whether the offer satisfies the criteria of "an offer" in a legal sense. Unless purchases for 8,000 GBP are generally negotiated in a chat window on a website then this kinda seems unlikely. Anyhow, let's assume it is an actual "offer".
If the offeror renegs on the contract, they may be liable to a claim from the buyer for the costs incurred by they've incurred as a result of the bogus purchase.
For example, if the buyer dispatched a truck to the vendors warehouse and on arrival they were told the sale wouldn't go through, then the vendor might be liable for the cost of the delivery truck for a few hours.
In reality, no one is going to bother making a legal claim for that small amount of money.
getting an ceo email address has never been easier
This makes me so happy. Based customer.
edit: sadly it wasn't like the original post made it sound like:
Code wasn't accepted. He copy/pasted his [fake, hallucinated by AI] code into the order comments section when he paid his deposit. He demanded the figures be adjusted for his discount code.
Bummer
But think of all of the money saved on customer service!
Where can I find that company? Asking for a friend.
. <---- Do you see this violin? It is what I am playing for you now.
if you're enough of a sucker to think AI can do your sales transactions for you, and dumb enough to go to reddit for legal advice, then maybe being a business owner isn't the ideal path for you
As per OOP's comments, the AI was only informational. It didn't actually create a discount of any kind:
Code wasn't accepted. He copy/pasted his code into the order comments section when he paid his deposit. He demanded the figures be adjusted for his discount code.
I'm not sure about the UK, but where I live, if the price is clearly a mistake, say 10x cheaper than it should be, the seller doesn't need to honour it.
Given this was pretty clearly adversarial prompting, it wouldn't stand up in court here.
"We did it, Patrick! We made a technological breakthrough!"
A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.
AI, in this case, refers to LLMs, GPT technology, and anything listed as "AI" meant to increase market valuations.
