You can still use the drive through if you like. Just mention your order code instead of telling them your whole order.
This seems like a great way to set prices according to how munch money they think you have or based on what you normally purchase. You might end up being charged more for the same items than they're charging the person who pulls in after you.
No discount on their shitty food is worth the intimate details of your life. They sure aren't collecting all that information because they want to make less money off of you. Stop giving them the valuable information they need to take more of your money.
That is exactly what happens if you don't use the app.
The mom-and-pop places that people who make a fuss about these kind of business practices tend to love do all the things that those people claim to hate - track their customers, try to figure out how much money they have, offer special treatment to particular customers, notice where else their customers spend their time - so I've never quite got what exactly people are objecting to. But hey, if you don't want the discount, more for me.
People always talk about apps "selling your data," and I can understand it in cases like Facebook (and Meta platforms in general) and Google, but in this case, how would it happen that McDonalds "makes millions off all the data you consent to giving them?"
That’s what the army of lawyers is for.
On the other hand, I also wonder if humans are too good at this task to make a big capital investment worth it. The employees who take your orders at McDonald’s aren’t standing around doing nothing, they’re wearing a headset and multitasking.
That's a math conclusion so lets show our work
Minimum wage is $16/hr in California + employee overheard so lets say $25/hr. With 3 people covering shifts at the drive thru, that's $219k per year.
Now for this system, lets look at off the shelf retail API prices to get a conservative guess. $0.006/min or $8.64 per 24 hours to listen & transcribe the customer, then add TTS + GPT-4o for $50 per million tokens in and out (thats way more than they need for a day). So that's $21k per year.
So back of the napkin math says it's 10x cheaper to implement this system. Unless you're referring to the corporate R&D budget.
There have been tons of articles recently about how rising fast food prices are causing people to consider it a "luxury", and a huge part of those rising prices are rising wage costs. Where I live all the McDonald's now have kiosks inside for orders and fewer cashiers. I absolutely think shaving labor costs is something all fast food joints are interested in.
I do agree with the other comment that drive-through workers multitask (scooping fries, filling drinks, etc.), so I'm not sure how much savings you'll get just from order automation.
Seems to work just fine the few times I’ve tried it, but as a native speaker of US English without much of regional accent I’m not exactly a tough test.
https://www.businessinsider.com/mcdonalds-ai-voice-order-tec...
>Under the partnership, IBM acquired McD Tech Labs, which McDonald's created after taking control of the AI speech company Apprente in 2019.
Probably also using the cheapest non-noise-cancelling microphones they could find at the time.
I asked the real human at the window whether they liked the AI system. They said no, it made far too many mistakes. In fact, the company was going to be uninstalling it soon because it was wasting too much time.
I can think of about a million reasons that might be, but one which particularly came to mind is that I live in an area where around 40% of people are Spanish speakers as their first language, and many of them speak no or very little English. The drive-thru employees would be completely capable of switching to Spanish (or summoning a Spanish-speaking employee) as needed, but according to the woman at the window, the AI couldn't do that. It also had issues with accents.
None of that is unsolvable, but it's disappointing that it apparently wasn't even thought of.
Why does everyone want to use AI to _replace_ humans?[0] This sounds exactly like a case where an "AI" transcribing a human's order ALONGSIDE an actual human would be a perfect case for optimising.
If the human mishears something for example, they can check what the AI interpreted, but still make their own decisions.
[0] We all know: Greed. AI's don't join unions and don't get overtime pay.
