If an LLM can’t be trusted with a fast food order, I can’t imagine what it is reliable enough for. I really was expecting this was the easy use case for the things.
It sounds like most orders still worked, so I guess we’ll see if other chains come to the same conclusion.
I can’t imagine what it is reliable enough for.
Apparently it’s good with knots.
🎶 There goes my hero 🎶
Hilarious, what a hero. Thank you king, this is the way to fight unsecure AI slop while they pretend its a legitimate reason to halve their workforce.
Computer can’t understand voices good enough so they think you say something different. For example standard voice to text has failed me for years. Voice control for years and years has still failed me. Using gps? fail every time because the computer or phone in this case thinks i say other words that sound nowhere the same. This is before and after AI. The biggest problem is voice recognition of words. Humans can separate sound and frequency better. Ear drums are better than microphones. Even in quite environments i have a voice that doesn’t get translated properly and fails more often than it doesn’t. So AI is going to glitch on a lot of voices too because the voice recognition part still works the same flawed way with flawed microphone technology.
“That’s a ‘you’ problem.” – Tech bros, probably
Since 2023, the fast-food chain has introduced the technology at over 500 locations in the US, with the aim of reducing mistakes and speeding up orders.
Bullshit, they just want to save money by replacing employees with AI.
Step 1: replace all your workers with AI Step 2: set new record for unemployment Step 3: most people can’t afford your product Step 4: increase the price to compensate for decreasing sales Step 5: blame Obama
They don’t even reduce cost from the savings. They charge the exact same amount until they want to increase prices.
This is a problem that requires zero AI.
A touchscreen kiosk is all you need to have the customer place their own order. You just need to have 2-3 ordering stations as they will be slower than trainer staff.
But shareholders demand AI
Also software voice recognition and using this for a shopping cart/order placement system (via phone line) has been a thing since the early 2000s, i know some mail order shops in Germany who used this.
Touchscreen kiosks are hard to reach through car windows.
Of course, the real solution is banning drive-throughs because they’re fucking terrible urbanism to begin with:
https://www.strongtowns.org/journal/2020/9/21/no-we-still-dont-need-drive-throughs
https://www.cnn.com/2023/06/24/business/drive-thru-fast-food-chick-fil-a-urban-planning
God damn. I was just about to invite you to the fuck cars community. In doing so, I needed the like with the !.
In trying to find it on the sidebar, I see you’ve found that community. As a mod.
Which means the whole basis for my comment is void before I even typed it!
On a similiar note…can you put the ! link in the sidebar?
I’m not sure which mobile clients have this feature but with a browser, you can just type
!and the first couple letters of the community name. It’ll start offering suggestions to choose from and narrow those down as you keep writing out the name. Making a selection fills in the rest of the link markdown for you.
Man, I never used fast food, or drive throughs as much as I have since I developed a mobility disorder. Last week I put a pickup order in at my local coffee shop out of habit, and couldn’t carry both my coffee and the breakfast sandwich to my car at the same time. Which sounds so stupid, but it took so much extra energy for both trips into the store that I was ready to go home and call it a day after that lol
I know the answer is “don’t get fast food and just eat at home”, but I’ve also been so tired after work/school that I’m not eating, and I dunno what the answer to that is either. My state isn’t a place where people think about how to care for their communities, and most of it has hours of highway between “cities”
I know the answer is “don’t get fast food and just eat at home”
No it’s not. Well, at least not in terms of urbanism/mobility, anyway; YMMV on your household budgeting.
The answer is that you shouldn’t have to get in a car at all between your home and your local coffee shop to begin with. It should be no more than a short walk (or wheelchair ride or whatever), door to door.
It’s a mile, and across an interstate exit, to my nearest bus stop. And I live in the only city in the enormous state that takes public transportation seriously.
I think I feel bad when I read articles like the ones you posted, before this I’d cross the distances and not think much of it because my last two cities didn’t have public transportation. Now I can’t cross fast enough to beat the crossing light, and it’s so incredibly unsafe if I fall. I feel like the problem, I guess
It’s a mile, and across an interstate exit, to my nearest bus stop. And I live in the only city in the enormous state that takes public transportation seriously.
Your first sentence contradicts your second. Clearly, there are zero cities in your enormous state that take public transportation seriously.
It’s crazy how most Americans have absolutely zero concept of just how bad even “good” infrastructure and city design in the US is, or how much better it could be.
They do not contradict each other. I’m certain there will be more stops as the city grows, because they keep improving it. I used to live in a city, in another state, with one of the best public transportation systems in the country, and they also kept improving that system to include the surrounding cities in other counties. Just because something isn’t perfect already does not mean we can’t take it seriously and strive for perfection
I forgot that US cars are so massive that this is a problem.
Agree on the urban planning part.
You could basically get around this with 4 programmers and an expectation around word choice. Voice to text has improved massively, you could just require them to say menu items with modifiers
Not that I disagree about drive throughs in concept, but this is just another problem easily solved by anything but raw LLMs
Completely right. People forget technology like Amazon’s Alexa (not an endorsenent) existed before LLMs. Speech recognotion just got better.
yes… it’s improved dramatically.
which is why 18,000 waters got ordered instead of 8 (or whatever.)
I actually don’t know why entering your own order hasn’t taken off more, come to think of it.
Chick-fil-A even went the other way, putting 4 high schoolers out in the weather for it. Cold, hot, raining, doesn’t matter.
With them I’m willing to just put it down to general crazyness.
That’s how my local Taco Bell is. If you don’t use your phone to place the order when you walk in there’s four kiosks you walk up to for placing your order. None of the people working are working the counter.
Most people have a hard enough time getting their car close enough to the window to get the food or to handle the card reader. I doubt many people would enjoy the toucn screen ordering as the would struggle tk get close enough and may have to lean way out of their window to order. The touch screen would also have to work reliably while wet or freezing cold.
Then you’ve got the issue of someone has to clean it and even with regular cleaning it has a huge potential to spread germs. Ultimately with all the disadvantages i think it would be too slow, too inconvenient, and add extra real estate to the drive thru.
Yeah, I guess you’d need some kind of arm or something to put the screen across. And then I guess you’d have to worry about it being damaged. Meanwhile, hiccups aside, an LLM is pretty much a drop-in solution.
This is not AI failing to do an easy job. This is “unskilled” labor doing complex and demanding work that cannot be duplicated by trillion dollar software.
I mean, unskilled just means minimal extra training is needed, not that it’s not complicated. Actual non-complicated jobs were automated last century in the West.
Tbh this is an incredibly easy fix, either cap the number of waters someone can order in software or have an override where a human takes over if an order is suspicious, there’s not an infinite number of ways to fuck with this.
Capping waters fixes that one specific issue but not the problem.
A suspicious order isn’t easy to define and no person who has ever participated in software development would underestimate the infinite ways a User can break software.
There are machine learning algorithms for anomaly detection though. They actually work decently well because exploits like this do in fact differ significantly from regular orders. Because they assume all anomalies are attempted exploits, their false negative rate is rather low while their false positive rate can be a bit higher.
Taco Bell has the capability to create a decently large training set from all recorded orders (which must all be valid and non-malicious) so they shouldn’t have too many issues developing this model.
If an anomaly is detected, make a human verify it is indeed an irregular order.
Surely if the person making the order sees 18,000 waters they would think, hold on this doesn’t seem right maybe I should ask the customer if they really want 18,000 waters?
The same applies for the ice cream with bacon on it which was mentioned in the article. I believe a lot of these could be resolved with a bit of common sense.
The same applies for the ice cream with bacon on it
Does it, though? Unlike the 18,000 waters, if I were working a drive through I wouldn’t even blink at an order for bacon ice cream. Heck, I might make a little extra to try it for myself!
Sure, but how do you distill this into a rule a computer can follow? “Suspicious” is not an objectively measurable thing that a program can just check against
Think the easiest way would be to collect order data for at least a good number of months if not a couple years and feed it in and use that as a baseline of what a typical human order looks like, anything that deviates too far from that baseline needs to be handled by a human until someone can validate it as a good order, though I imagine you could get false positives for new menu items unless you set a reasonable instruction for items that have never appeared in the dataset before.
Sure, in the most extreme cases it would be obvious to the crew. But simply making mistakes at a higher rate than humans will result in a lot of unhappy customers.
there is an incredibly finite number of ways to mess with this, they just need a button to send a report to the engineers with how they got messed with and eventually they’ll have a complete list. I really doubt it’d take long to iron out the vast majority of ways that can be thought of.
A QA engineer walks into a bar and orders a beer.
She orders 2 beers.
She orders 0 beers.
She orders -1 beers.
She orders a lizard.
She orders a NULLPTR.
She tries to leave without paying.
Satisfied, she declares the bar ready for business. The first customer comes in an orders a beer. They finish their drink, and then ask where the bathroom is.
The bar explodes.
This isn’t something you can input any text into, it’s fixed, that joke doesn’t apply, you can’t do an sql injection here.
Close one, a joke was related to but not a perfect match for the present situation. Something terrible could have happened like… Uh…
Let me get back to you on that.
Sounds like you’ve never programmed before.
If what I was saying was false, there would be the same issues with the touchscreen ones.
that’s what happens 99% of the time. It’s kinda been a trend on the anti clanker side of TikTok, just order a large amount of stuff so a human takes over and actually helps you
Why can’t a trillion dollar AI say “Sir, that’s not reasonable”?
Because they train these to be your cheerleader, not some back talking reasonable person.
The point is that loopholes in software will always exist that lead to unexpected outcomes.
In software in general? Sure, here? I really doubt it
It actually crashed, per the article. Presumably because it failed a sanity check.
You know what would reduce mistakes and speed up orders? Replacing the god-awful audio equipment that every drive thru seems to use. It wouldn’t cost much to install good audio equipment and it would quickly pay for itself in time savings since people could hear each other more clearly.
Can’t find a source for this but I remember reading that the audio is intentionally bad because it subconsciously makes people more inclined/pressured to purchasing add-ons or upsells.
What it makes me inclined to do is never use the drive-thru, personally.
Years ago, I was ripped to shreds for saying AI would be a disaster.
it’s not ‘ai’, it’s just a poorly trained voice recognition system that’s trying to decipher any random person’s voice. voice rec has a difficult-enough time when trained on a single person in a professional setting (lawyer, doctor, etc)–which is work we used to do here. it’s going to absolutely fail in a ‘drive thru’ environment.
it’s not ‘ai’, it’s just a poorly trained voice recognition system that’s trying to decipher any random person’s voice.
I’m baffled that you can say “It’s not ‘AI,’ it’s a machine learning powered speech to text system” with a straight face.
Even if we were to agree that ML-powered speech to text isn’t AI (and I don’t agree to that premise, for the record), there’s still the matter of processing the transcription to transform it into something that can be understood by the point of sale system - aka natural language processing. And while that NLP could be implemented without use of an LLM, given LLM’s current level of hype and the ease with which they can be shoved into any given product, I wouldn’t bet on Taco Bell execs approving such an approach, much less asking for it.
“It’s not AI, it’s just <multiple things that all fall under the category of Artificial Intelligence>”.
AI is a huge field of computer science. It’s not the one tiny narrow definition of artificial general intelligence like HAL 9000 or Skynet or Detroit Become Human.
It’s not AI unless it’s running on a cluster of H200s. Anything else is just sparkling text prediction.
