Hi, I work as a consultant around AI, and I want everyone here to know: every company with a brick-and-mortar wants to do this. The barriers are (1) tech stack, (2) use case, and (3) ethics, in that order.
It's not just facial recognition in stores: it's tying face, license plate, credit card, and phone app into a single customer view; it's pricing based on your past (and current!) behavior; and it's data consortiums to tie together data across companies (ever wonder why everyone asks for your phone #? It's traditionally the primary key in the big-*ss joins).
The expectation of privacy we have online is coming to the real world, make no mistake. I wish I had a call to action here but... keep yelling, I guess? I don't have a solution.
If your real name is discernible from your HN profile, and you own a car, I could likely get your license plate. If you live in a big city, I could search your license plate and find locations where you were. Alternatively, I could find your address, then search for license plates that drove near your house. It's not a perfect system but it has millions of records. And there must be other correlated databases that I don't know about.
The only things we can do is use fake/different numbers at stores, use prepaid or business debit cards and have burner iPhones and rooted Androids. Register your license plate in corporations names, and don't put your utilities in your real name. If we just decorrelate the data, and falsify it, then we could have a chance.
You can put bluetooth beacons throughout a store. If you have a loyalty app on ios, it will inform of your precise location. On Android I suspect it is much worse and you might not need anything. Google probably collects each contact. I don't know what gets passed on, but some edison beacons have URLs.
When you go to the registers there will be beacons there and it can correlate it all to the identity of the purchaser.
If you have wifi on and I believe if you have connected to a private ssid, your phone will broadcast requests for it. You can passively listen to phones and look these up in online ssid to location databases.
You can turn off bluetooth and wifi (although apple has dark patterns wrt to this and android? maybe you can turn them off?)
I was under the impression that multiple "towing and recovery" companies drive around all the strip malls every day with lprs and give each other access to the data in order to help each other find cars to steal back for the lien holders..
In some cases the malls themselves allowed stationary cams to be setup at the entrances and used for these purposes - I think an article (in wired?) about a year ago shed light on one in california that maybe changed the cam feed relationship after pointing to a 'terms were posted' defense.
I try to remind people I see that are okay with this tech being deployed at the airport and other places - that the state spending a bunch of money 'to keep an eye out for stolen cars' is just an excuse to grab data that can later be looked at - and is sometimes given to repo men so they can take cars from people who are late on payments.
It is registered in almost all systems already and you'll get the discounts. They get nothing (except the knowledge that apparently Jenny buys everything). Plus, nowadays most people working at checkout don't even know the significance of that number and don't bat an eye when you give it
This number is also almost always good for like a dollar off per gallon of gas at stations that are tied into the grocery loyalty program. Lots of SafeWay's have their own stations for example.
Corporate data miners care if you reuse someone else's registered phone number.
The cashier doesn't. Their incentive is to keep the checkout line moving. Sometimes they will even offer you a card at the register to get you a discount (not sure why).
The manager's incentive is to keep customers content and returning.
BTW, what is the significance of that specific phone number?
> Sometimes they will even offer you a card at the register to get you a discount (not sure why).
Because they are told to do so, to get as many customers as possible use the card. My wife automatically answers "no, we haven't got any and are not interested, thanks".
> BTW, what is the significance of that specific phone number?
How does having all the customers use the cashier's discount card help the company at all? I've heard the opposite in many stores, where they prohibit the cashier from doing that since it's giving away money (and loyalty points to the cashier) without collecting any data.
On a side note, the phone system has a reserved number range (555-XXXX) for fictitious numbers for TV/movies/etc. to use. This song didn't use that but instead used a real number. When the song came out in 1981, people who had this number got so many phone calls they had to change their number. I'm not sure, but phone companies may have retired the number because it has become impractical to use.
Similar to a social security number[1] that was publicly distributed with wallets at a large retailer in the late 1930s which became the most commonly reused / misused social security number. It also caused the owner lots of grief.
The USA really needs a new amendment to protect people from digital surveillance both from governments and corporations (businesses). In olden times there was a limit with mobility and technology. That is no longer the case. A similar motion in the United Nations would also be useful but with Russia and China 2 of the 5 members of the SC I don't see that ever happening.
Now that facemasks are normalized in the American public, surely this tech is pretty inert now? I imagine people will start wearing their masks for any cold or flu, not just COVID, like other countries have been doing for years. Any would be shoplift is now socially allowed to conceal their face at all times.
And if you weren’t a shop lift, it sounds like taking public transit, walking, or biking, and paying with cash will be king for the privacy conscious. At least this will force a few people into better mobility habits.
>Now that facemasks are normalized in the American public, surely this tech is pretty inert now?
No, facial recognition is based on the distance of points on your face (nose, center of eyes, eyebrows, etc). A mask only covers the lower half of your face. It may make facial recognition slightly less accurate but it still works. Casinos have been using facial recognition for some time to detect individuals who have been banned. These banned individuals have gone as far as hiring professional make-up artists to change their appearances but facial recognition still works due to the fact that certain points on the face (e.g. distance between eyes) cannot be altered. The accuracy is reduced but there's still a large enough confidence interval to cause an intervention by casino security.
In that case, are masks + sunglasses enough? Or should a hat that covers the ears be thrown in as well? I think that combination should still pass as fairly normal, especially in sunny locales.
Yeah this is why we need to make it normal to wear sun glasses and trench coats and baseball caps. Both a fashion statement and a rejection of surveillance everywhere. Also bitcoin and cash should be as normal as popping out a credit card
>Yeah this is why we need to make it normal to wear sun glasses and trench coats and baseball caps
I think we could also simply pass laws against digital surveillance instead of forcing everyone to dress like a mix of Inspector Gadget and a sex offender
> Now that facemasks are normalized in the American public, surely this tech is pretty inert now?
Only if you're wearing sunglasses and a hat. Even then there are ways to get at people, such as gait analysis, or your cellphone tattling on you via bluetooth or wifi.
I'm going to go out on a limb and say the anti-mask campaigns were started by parties that have vested interests in facial tracking (I don't have any evidence to support this assumption).
Coupons, upselling, personalized-proactive-ads. JC Penny showed people would rather pay MORE with a 70% discount than less with no discount. Jack up the price, get everyone to download your sh*tty little spyware app, and give everyone the thrill of "I got a free doodad!"
One business that's already great at this is Target. I've observed through personal experience, they will send a coupon for a really great price on items you normally buy like diapers or formula, if you have kids, to try to get you back in the store. They're not sending that diaper coupon to every customer.
Target has less people and they don't play music there, which I appreciate. Our local Target is quieter than the local library. Nevertheless, I attempt to shop at local stores and I avoid both Walmart and Target as much as possible.
Maybe basically, but at least in my area there are far fewer instances of people wearing pajamas in Target and it's much closer to where I live. I don't think you can draw 100% equivalence between the two stores enough to support an assertion of an advertising-vulnerable demographic.
lol, out here in los angeles, they've been remodeling targets to look like walmarts - like using polished concrete floors and skinnier, cheaper looking checkout stands with flimsy conveyer belts. The most irritating thing was when they replaced the target cafe with.. nothing. Some stores have a starbucks, others have.. nothing.
Oh yeah, and Target pharmacy was sweet. CVS pharmacy inside target isn't as good.
That said, they are still somewhat faster and have way better hours. The walmart where I live is only open until 10pm and it has 10+ minute checkout lines. The target is open until midnight.
OTOH, online ordering: Walmart is often better than amazon. Target's online ordering sucks.
The same way coupons work in drug stores already; there's a stated price and then there's the price with SuperSaver™ membership, which you have to pull out of their app and could be subject to any number of personalizations, real-time or not.
Every time I check out at a Kroger store I tell the cashier that I would like to get a membership card. They don't require you to give a phone number until you register the card after the checkout - which I don't do because I already got the benefit of the card at checkout. This won't work at stores that require phone number first, and still leaves open credit card-based tracking - for that you actually lose the few % cash back if you want to use cash instead.
I've been using the same un-registered Smith's (Kroger by another name) card for 5 years now. I get gas points and everything. Now, do they know a lot about me based on my credit card? Yes, but it's a start...
I have one of those too.. but only used cash with it. I have heard that if you use that and a CC or debit card they link the data - even if just once - and can link the past data as well.
I also have my wifi turned on my phone so they can't correlate the 'burner card' with that.. if I don't have cash I use a different discount card.
Of course if they scan the face and correlate that way - meh. For that reason I only use cash at Target - but not sure how long that will be enough to keep from data points being put together there or anywhere at this rate.
Every time I check out at Kroger (or similar membership-based store) I tell the cashier that I don't want a membership but I do want the listed sale price. They always just scan their own card.
CVS started to refuse to scan the 'house' card as of about a year ago. and they try to collect your phone number when they hand you a card for the first time.
But both of those prices are always shown and you don't usually have to be a member of anything to get them as the clerk will just use the generic code. At most self checkouts there is an option for forgotten card which lets you bypass the need for an account.
There already are private prices -- the result of targeted coupons that certain customers get. Those coupons are tied to the SuperSaver(tm) account and aren't transferable.
Right, those won't go away. These offers will likely be in addition to those and because they're more targeted they will be higher value. CVS already has specific offers that you can only get in their app with your specific membership, Target uses their Circle membership to accomplish it.
The medium to deliver this customer experience are out there right now, but they're all working on adding signals to inform which offers you get.
Ever seen the digital price tags at Kohl's? I can imagine a scenario where you walk in and the prices change!
More likely, though, it would be pricing things differently at stores based on demographics. Regional grocery store chains do this - We have Stop & Shop here in New England, and if I go to a store closer to an urban area, the prices are lower on some items and higher on other items.
I am in Wales, we don't have such things here. I think I'd be reluctant to shop anywhere that did this. What would be the point of working hard to earn more money if all the prices went up around you so you were no better off than before?
Companies use this data to raise the price of items that are mostly bought by people with a lower sensitivity to price.
A famous example is Tesco in the UK that worked out they could double the prices of certain cheeses with no impact on sales volume, because the people that bought those cheeses mostly bought the most expensive brands in other categories (they were probably higher income people).
I get a piece of mail from my local Meijer (midwest grocery/department store, think Walmart) every week on Thursday, a couple days before my grocery run on Saturday.
It contains about a dozen coupons for the exact items I typically buy:
- Store-brand milk, 2%
- Honey bunches of oats with almonds cereal
- Bone-in chicken breasts, family pack
- Noosa strawberry-rhubarb yogurt
- Old spice aluminum-free deodorant, if it's been a few weeks since my last purchase
etc. etc. etc. It's not the generic "all Pepsi/Nestle products on sale" mailers. It's the exact stuff I've bought in the past. And yes, I admit I'm a creature of habit. There's little $4/hour coupon clipping or recipe selection required (I hate that), I can pretty much shop my normal list and then swipe the coupons across the scanner. And tell you what, if I skip a week at Meijer and go to Trader Joe's instead, there's likely a few more coupons for a little deeper discount the next week.
It's more than a little creepy. I have no idea how they got my home address, maybe scanned from my driver's license when I bought alcohol? And I don't use a store membership, they must be just tying my debit card number to my shopping habits.
If you use a credit card to pay and they datashare with another company where you used the same credit card to pay and the other company has your address, Meijer could have gotten it from them.
I have a walmart.com account - they add my in-store purchases to my Walmart.com history when I pay with a credit card I used at walmart.com.
Aren't there already some experimental stores where you have to scan an item with your phone in order to find out the price? If the app is linked to your identity, then you could see a different price than others based on your demographics and past purchasing behavior.
Shop at local businesses as much as possible. In my area, this is somewhat viable. I'm not holding my breath on much of the populace actually doing this though. If we haven't boycotted Walmart by now, we're not ever going to.
Unfortunately a lot of local business seems to be "here, let me order that for you (more slowly, with a mark up instead of a return policy)."
Most people I know genuinely want to support local business, but simply can't because local businesses do not provide them any additional value for the often higher price.
A functioning local economy is a nice perk for the money.
The essence of the problems with both the online surveillance state we've created and the burgeoning meatspace one we're sprinting towards is that people (in America, at least) have effectively ignored the knock-on effects of their purchases for the last, I don't know, 50 years or so. The cost of a functioning downtown is a couple bucks more, the cost of a local economy that's worth a damn is a couple bucks more, the cost of knowing where your products came from and that there's not a dead body or slave labor in their past is a couple bucks more.
I'm not gonna judge someone who says "I can't afford a couple bucks more," but if you're on this site and working in this industry and you're mad that Rite Aid is stalking you, pay a couple bucks more.
I think that's it in a nutshell and it's something I've decided to do...pay a couple bucks more. Or, in a lot of cases, re-evaluate the purchase and pass altogether. Many of the things we pick up from WalMart aren't really necessary.
I think it's more of a question of what the economy functions on. We don't need people tied up in waiting for customers' orders and sending away for them over several weeks. It isn't productive. A local economy that's based around a similar scheme is only accessible to those with disposable income, and will tend towards items that are superfluous to begin with (if I need something, I find it as quickly as possible; I'll only wait if I don't care that much). This strikes me as a situation lacking in sustainability.
If part of the value proposition includes not being stalked, that becomes more compelling. However, it's often the small guys that explicitly try to build their phone number database. The confusion when I ask if I can just "skip it" tell the story. Either way I truly hope the situation improves.
Having relationships with real-world people is really nothing like a faceless corporation compiling a database on you and I think you probably know that.
Edit: I may have misunderstood the tone of your comment. Are you implying it's bad that people at your local store might know who you are? I should have my coffee.
Yup - then the whole world won't be able to automatically know you. It seems like a decent trade-off. We've already been using that model for many, many years.
I assumed that they'd just link it up with credit cards even if I refuse to give my phone number. Or, maybe they'll invent some common encrypted face fingerprint with which they can identify customers across companies.
Most of the small indi-shop ipad POS devices do this already. If I swipe my card at one, I get an email automatically with my reciept. At some point in the last few years I must have put my email address in one and obviously it's shared.
Credit cards are harder, because people switch those. I have had the same phone number for over a decade. This way, they keep historical data in the same profile.
May I ask what you mean when you list the use case as a barrier? Just that they know they could do something, but exactly what to do and how to finish it is the hard part?
This would almost certainly raise prices across the board, which would end up hurting poor people.
Price targeting can be (and often is) a useful and mutually beneficial practice. Obviously it can be taken too far as we're seeing here. I think enforcing privacy rights of consumers would be a better legislative goal than uniform pricing.
But price targeting effectively increases information asymmetry. It would be a truly bizarre turn of events when the party on the wrong side of that asymmetry would benefit from making the asymmetry worse.
The price of a loaf of bread will probably just rise, for each individual, until everyone is reduced to different shades of poor, no matter how much they earn.
I understand the vision for this technology involves marketing information, but this article is mostly about security and protection against theft. If it is useful for that alone, that would make it an invaluable tool. The regulations to protect consumer privacy should be careful to keep that distinction in mind. We all pay the costs of theft.
Let's give everyone a unique number and a barcode on their wrist. It's useful for "security and protection against theft". This has been tried before. There is a reason gun rights advocates have lobbied Congress to prevent the FBI and ATF from keeping digital databases on gun owners from the applications created when buying a gun.
Your digital facial likeness will almost certainly be collected, collated with other data points, then both data mined and resold to data brokers. One of them will no doubt be dumb enough to leave their DB open to the wide internet and your data will be leaked and you will have no recourse because you can't prove damages in the US legal system and you aren't the customer of the data broker.
I mean, I carry my phone around in my pocket, so I kinda expect at least big Google is watching me. And probably the cell network. And probably the government, passively.
But RITE AID? What the hell, a fucking pharmacy? I mean: the customer is typically just buying whatever their doctor prescribes them. I guess they have aisles of cosmetics and cold medicine and stuff, but still, this seems remarkable for its banality and at the same time invasiveness. It's the worst kind of evil: stupid evil.
I switched to a PinePhone recently and have another I'm putting Andoird+miniG on, so I have at least one thing out of the way.
I never sign up for loyalty programs (phone number), but I still pay with a debt card (I'm not sure if store can track people without violating card ISO agreements, but I suspect they can with hashed numbers), so I'm not full RMS/cash only.
As far as shrink prevention .. eh. I don't have a big issue with that. I've watch a woman just walk out with shit in an inner city store. Dude in line was like "She just walked out with two wine bottles," and the cashier was like, "She does that all the time. If the guard is here, he'll stop her. I'm not paid enough."
The article makes a big deal about certain neighborhoods. I'm pretty sure they just selected the stores with the highest reported shoplifting losses. Trying to read more into it is giving the data more intrinsic intelligence or prejudice than it has.
I think it's a factor of these systems being cheap. Cheap software, cheap cameras, cheap storage, cheap compute. It's a shame we don't have a better use for all that stuff.
I mean, one thing we all can certainly be doing is making a concerted effort to shop more at our local independent businesses. Plus, this comes with the added benefit that it keeps money circulating in our communities for longer!
This sounds good in theory, but my corner store sells a bottle of soda for $3 that I can get at Safeway for $1. Or a bag of chips for $2 that I can get 2 for a dollar at Safeway or Target. Sunscreen for $12 instead of the $6 (or as low as $3 on sale/clearance) at a big box store. And most of the stuff I want they just don’t have at all. It’s not like a 10% “shop local” tax, it’s really unjustifiable the insane price differences. I’d be better off just donating to a local charity.
And to the original point of reducing tracking, for Safeway/etc rewards numbers just use random local phone numbers and pay cash if that’s your priority. Better than paying more than double price for limited selection.
That's a bit of an apples-to-oranges comparison though. Things are more expensive at the bodega partially for convenience after all, they aren't directly competing with Safeway.
In a lot of cases there is an in between. I live near an independent full service grocery store (they have 2 locations) which isn't the cheapest in town and isn't the most expensive. They have a good selection. If they are 10% more expensive, I'm happy to support them.
If they are hiking prices up this much because of convenience, then they really shouldn't complain when people take a little bit of extra effort to go to the nearest big store.
The real value proposition of a bodega/corner store is that it it's on your corner, or the one 1/2 a block away. They used to have longer hours too, less so now that grocery stores open late or 24hr. So it isn't the place you went for a big grocery shop, but the place you dropped by when you forgot something or ran out of milk.
They've always been more expensive than grocery stores, doesn't matter if they are local independents or big chains.
As @ska pointed out, I'm not really talking about buying all of your groceries at the corner store at a 3x markup. In every city I've ever lived in, I've had the option of taking my business to an independently owned (grocery|hardware|pharmacy|etc.) store where the "10% shop local tax" more or less applied. To me this seems like a fair exchange for the peace of mind I get from knowing that I'm not being targeted by government-grade surveillance tech when I walk through the door, and that the employees I interact with are being paid a living wage. YMMV depending where you live but I hardly believe this is a phenomenon isolated only to the places where I've lived.
Also, if your local shop doesn't carry an item you want, you can always speak with management and ask for them to stock it :)
At the end of the day, I can rest easy knowing that all of this is to shove products in my face that I still don’t want or have any use for.
All of the analytics of the internet, all of this data collection and machine learning - all to show my a picture of a $69 leather jacket. Or a Muse headband. Or a Peloton bike.
I’m not a consumer, these tricks don’t work on me.
Yes, every once in a while there is a product that does not need advertising or aggressive marketing, where it's value is immediately and plainly obvious to everyone watching.
Buying one of these things does not enter you into the consumer-product-churn-rat-race.
I'm in the same boat, i'm not a consumer as they'd want us to be, I don't buy into any bullshit marketing, I never buy advertised stuff that passes my add blockers, I don't watch TV and when I unavoidably see billboards outside I make a mental note not to buy those products specifically because all the money spent on adverting will be paid by the customer in the end and Im not that sucker. I think advertising is toxic and obnoxious, but that's not everybody's opinion. I have some acquaintances who don't bother to avoid advertising and might as well find them fun even if they've seen them hundreds of times. I have no hope for these people, but they are are model consumers and somebody's plan is working accordingly
I always wonder if they really get much value out of all this surveillance or if they just hope to get value.
It seems that a lot of companies are these days are trying to hack their customers instead of treating them right. In the long run I don't this will work. I already am getting a very bad taste dealing with a lot of companies because they seem to view me, the customer, as something to extract money from instead of a customer to be respected. You notice that for example with a lot of search results where the company places stuff they want to sell in the results instead of allowing me to search for what I need.
Either we are going into a very dark future or there will be massive pushback. Not sure which way it will go.
Even companies like Google seem incapable of using all the information they gather in a very helpful way.
There are some small ml cases that seem to work ok, like suggestions for message responses.
But YouTube video suggestions for example seem so facile. If I click a few videos about today's news, then every video suggestion quickly becomes similar content. Even though I am subscribed to a dozen niche channels and may have watched a bunch of science and engineering videos, I don't get any suggestions related to that unless I directly seek them out. After which, all my suggestions turn to engineering videos.
A few weeks ago I started getting youtube video suggestions for religious Thai videos, which I have never shown any interest in. I ignored it. And it kept happening! I spend most of my time watching educational/tech review content, so I was worried my account got compromised. Changed passwords, and still the Thai videos kept getting recommended.
After a while I finally realized what happened: I would keep an ISS live-feed running in the background while I did housework. It was nice to just see where it was. Turns out that video was being hosted by a channel connected to several of these other Thai channels, and youtube assumed I would enjoy the content.
They still come up occasionally, but I've clicked "Not Interested" enough times that I think they got the idea.
I wonder if those Thai religious videos were spending a lot of money trying to get your specific demographic, so youtube showed it to you even though it was evident you had zero interest.
there's some kind of mob rule going on suggestion wise that's pretty annoying. What are some gem sci/engr channels you've discovered if you don't mind sharing.
Isn't this what they are optimizing for though? I don't work at Google so I don't know and I block ads so I am not engaging with their revenue generation (at least in the most direct sense) but the idea seems to be if you are watching X content you then want to see more of X and it drives total engagement time. I think it sucks and is part of general trend of not being able to find diverse and cool content on the web..but hey Google
Didnt Orbitz (or one of the travel sites) get caught charging customers more money per ticket if they were on a Macbook vs any other computer? [1] This type of data collection is the start of individual pricing based on what stores believe you will pay.
I don’t believe they were offering the same hotel room at different prices but rather more pricier options to different users. From the WSJ article that is the source of the CNET one:
“ Orbitz executives confirmed that the company is experimenting with showing different hotel offers to Mac and PC visitors, but said the company isn't showing the same room to different users at different prices. They also pointed out that users can opt to rank results by price.”
Stuff like this annoys the hell out of me. They should offer the same things at the same prices to all people. Same in Us healthcare. A procedure will has a several hundred percent price variation depending on insurance and other factors of the patient.
> It seems that a lot of companies are these days are trying to hack their customers instead of treating them right.
This is true, and frankly really well put. But I think it has its limits. Rite Aid and similar stores have an extremely tenuous "relationship" with their customers already. The pharmacy in the back might, but the overwhelming bulk of their revenue comes from the snacks and sundries they sell to people walking by, and frankly those people don't care any more about the store than it does about them.
I mean, it's cynical, but there's social value in having Just Another Faceless Junk Store on every corner. We don't want to be "treated right" as an independent variable, we want the drink cooler and sunscreen to be trivially available and we'll walk into whatever door we see.
In that world, "hacking your customers" is the only feasible path to growth. I share the upthread concern as to whether this is effective, I just don't see a way for these stores to avoid going down that path if it does work.
There's something horrific about running a society in such a way that the Just Another Faceless Junk Store on every corner is so worried about chasing growth that it decides to "hack its customers instead of treating them right".
"Hacking your customers" may not be a feasible path to growth if the customers don't like, especially they already lack any kind of loyalty to the business. I used to pop in there now and again because there's a location next to my house, but I won't be going back.
> In that world, "hacking your customers" is the only feasible path to growth. I share the upthread concern as to whether this is effective, I just don't see a way for these stores to avoid going down that path if it does work.
Maybe this is a reason to reorient our economy to focus on the well-being of its citizens as opposed to efficiency improvements in their monetization. If growth in sales/revenue/whatever wasn’t the goal, then the economy could focus on providing necessary services instead of “hacking” customers in order to sell more junk. The economy would be focused around giving people what they need as opposed to maximizing consumption.
>The continued investment implies that the market has spoken: Your data has value.
>Your data has value.
Does it. Harbor Freight already knows it's customers are cheapskates and Whole Foods already knows it's customers barely look at the prices? Backing that up with data is nice but how much value does it really have?
FOMO is a big thing at the corporate management level. A million here, a million there to run pilot programs and do partial roll-outs seems like cheap insurance against getting blindsided by something you wrote off as a fad.
> Does it. Harbor Freight already knows it's customers are cheapskates
Actually, that's a great example of a company that strategically prices items differently for different customers. Some of their customers are cheapskates and hit up hfqpdb.com and do the math to find the best coupon for every item they buy. Some just use the coupons in the flyer they get in the mail. Some of their customers pay the shelf price. The price essentially adjust to how cheap that particular customer is based on how much time they're willing to spend looking for coupons. Customers who don't care pay an inflated price.
I would bet a large portion of their profit comes from the people in the middle -- who were lured into the store with a coupon and end up buying a bunch of other stuff at inflated prices. I could totally see them being very interested at identifying who those customers are and how to lure them into the store.
==Harbor Freight already knows it's customers are cheapskates and Whole Foods already knows it's customers barely look at the prices?==
If this is how you think pricing decisions are made, I don't know what to say.
On the one hand, we have the lived experience of increasing surveillance and targeting, an investigative article discussing it, and long, documented history of data investments from large corporations.
On the other hand, we have speculation that it's all FOMO.
You don't run pilot programs and eight-year roll-outs of things you aren't taking seriously.
"All the smart guys are spending money on it, it must be valuable!"
You could use this logic to invest in real estate in 2007. Or you would've been eager to invest in WeWork had their IPO not been fucked. You would also believe self-driving cars are a good idea, but I'd bet my life savings against it knowing what I know.
Don't assume you're playing the same game as them when you can't actually provide any data that what they do works. Where is the actual "X dollars of surveillance turns into Y dollars return" data points? You're just providing speculation no better than the person you're arguing with, and honestly your argument is less compelling.
How much is the potential of loss prevention anyways? I've seen tangible effects in store where certain products are behind glass shelves and you must get permission to purchase one and then it goes in this funky plastic box you take with you to checkout, sometimes right around the corner by the glass cage. I'm assuming theft is growing to have responses like that enacted which are incredibly inconveniencing especially when there's no affiliate nearby.
Knowing I'll burn up a bunch of time waiting for one of those lockdowns to be opened I usually order online and pickup in store those items I know are locked up.
It’s probably more helpful to think about it at the margin: starting from some current point equilibrium for some spend $X what is the incremental lift $Y? Because that lens probably more accurately reflects a real purchase decision.
Harbor frieght sells the same shit for years at a time. Their sales are always rotating. As a customer I basically have enough info from 3 months of their ads that I have 0 fomo anymore and also I have a good idea of what the price on an item should be. So essentialy all of their targeting taught me how to maximize them.
Also due to any lack of slack in their product line, I don't have anything I'm missing out on.
So basically, sell me latex gloves for <$7 or I'm not coming in. Magnetic screw trays for $.99 or the bigger ones for $1.99 or not coming in. I can wait literal years for that motorcycle lift to drop to $250 instead of $475.
I guess that depends on how you define loyal. If I need a consumable tool I'll check there and amazon. Other than that I never go there. So loyal I guess?
yes, loyal doesn't mean exclusive, just that it's on your short list, which is immensely better for a business than not being on the short list (exclusivity is an unrealstic expectation anyway).
> The continued investment implies that the market has spoken: Your data has value.
That's a dangerous assumption to make, even irresponsible post-2008.
It's also a fallacy which gets used by scammers often.
For example, some in the spam industry relied on that fallacy to sell malvertising schemes to rubes. Because somebody must be making money off spam, right? But the scam is selling the scheme/software to enough spammers to extract a nice living before each of them predictably jumps ship. So the answer could be, "no, they aren't making money," or, "maybe, but you won't be," and the incentive to sell the systems is still there.
Data may have value, but continued investment in it is certainly not enough on its own to elucidate what that value is.
Grocery stores can sell client data to their suppliers and use it to negotiate with suppliers. They also monetize it by sending you (supposedly) well-timed coupons.
this is a huge money stream, and for retail they don't have many other options-- amazon and online delivery is eating their lunch.
Having worked on customer-tracking systems for several companies and seen executive strategy first hand and compiled reports to assess the effectiveness of these types of programs I have witnessed clear evidence that what you are saying is NOT true. Yes, there could theoretically be value to this type of data but trying to explain this to VPs is pearls before swine.
"They wouldn't do it if it didn't work" should be added to lists of logical fallacies. People do dumb stuff and make mistakes all the time, but groups of people are collectively about as clever as a slime mold.
To assume unquestioningly the intelligence of our rulers during the Trump era is just, wow. I don't know what world you live in, but in mine, the worst people on the planet are the ones at the wheel.
> "They wouldn't do it if it didn't work" should be added to lists of logical fallacies.
Exactly. Other reasons for doing something include:
- Thinking that it works, when it doesn't - in particular, reality is messy, and you can often get away believing in bullshit, as long as your peers believe in the same bullshit too.
- It not working for the company in general, but working for them individually to scam their own employer or boost their careers (related: CV-driven development in software).
- It not working, but the profit is made off making you believe they do it because it works.
speaking of fallacies, surely there's a middle ground where our "rulers" are not all geniuses (plutocracy vs meritocracy), but that on average companies pursue things that work over things that don't (survivorship bias).
and groups of people can be smart crowds or dumb (slime) mobs, depending on the systemic conditions.
The correctness of that conjecture depends on what you're averaging over. For example, racism and sexism. Those apparently can persist in the workplace, on average, despite the drag on profits. However, I'd agree that for the average topic, the market does a pretty good job of optimizing behavior.
that's more of a frame of reference issue though. in cases like racism/sexism, available profits on average do accrue to the people wielding power for that purpose, whether it's good for profits overall.
==To assume unquestioningly the intelligence of our rulers during the Trump era is just, wow.==
Who did this? I laid out the case that modern large corporations (no need to bring politics into it) are some of the most effective moneymaking machines in history. We can observe them investing heavily into customer surveillance/tracking. We can also observe the primary players in this industry thriving from a revenue and market cap perspective (Google, Facebook, Amazon).
The hypothesis is that both main street (through revenues) and wall street (through outrageous P/E ratios) are signaling that the data being collected does, in fact, have value.
At first I wasn't going to reply because you are clearly not discussing in good faith (you neglected to address my evidence to the contrary or the flaw in your reasoning but instead focused on my random afterthought.) But others might read this so I decided to elaborate a bit.
I was once misinformed with the belief that business' primary goal is to make money. But when you have the experience of explaining data to a CEO that clearly shows that their customer surveillance has a negative ROI and that the conclusions they are drawing from the data are not warranted. (For example, if you want to claim that your initiative increased sales, you need to compare an equivalent data point before and after the initiative, not just afterward.) If you say this and then look into the cold, dead eyes of a sociopath you'll know how foolish you were to ever assume the point of tracking customer data is to make money.
I felt dumb ever believing that business leaders care directly about money. I knew that to be false in high school when they made me read 1984. The lesson that O'Brien teaches Winston is one that we clearly have not learned.
> The object of persecution is persecution. The object of torture is torture. The object of power is power.
We can add to that list "The object of unrelenting invasive surveillance is unrelenting invasive surveillance."
I think you are misunderstanding my point, so apologies for being unclear. I am not questioning the actual validity of the data collection or ROI of the projects themselves. I am saying that the market can award value to companies who just signal they are making efforts to "digitally transform" or "invest in data assets" or "monetize data". It doesn't matter if they accomplish them tomorrow, the market is giving value today.
Thanks for sharing your CEO experience, that is not a data point I have encountered. I do think money and power are closely linked, although maybe not as closely as I assumed.
==I felt dumb ever believing that business leaders care directly about money.==
Today, what do you believe they care directly about? Power? Ego?
Technologies with similar capabilities are being rolled out en masse throughout retailers across America. I've worked with a regional grocer supporting one of these projects. There are turnkey systems from several vendors, but the general concept is similar. Cameras have full coverage of the retail floorspace from multiple angles. The systems can track an individual throughout the store, and they can observe someone pulling items off the shelf.
This is used for a couple things. One, if customer pulls product A off the shelf instead of product B right next to it, the manufacturer of product B can purchase the ability to have their coupon print out at checkout to be handed to somebody who is proven to have an interest in the category. Alternately, it can be used with self-checkout systems to confirm that the shopper actually scanned product A, or to alert security to check the shopper's receipt and cart at exit.
Amazon is offering this sort of technology for checkout-less stores, but similar offerings for loss-prevention and targeted marketing are online and available.
It would be trivial to add face recognition to this stack, and I strongly suspect it is in fact offered if you ask for it. The use case there would be focused around the loss prevention aspect, and that's a huge financial driver for implementing these systems.
Another interesting feature of this market is that the websites for the products themselves don't really advertise the full functionality of the product. You won't see the scale of the offering until you get to the in-person sales demo and slide decks. It feels like the vendors know how the technology could be perceived.
> One, if customer pulls product A off the shelf instead of product B right next to it, the manufacturer of product B can purchase the ability to have their coupon print out at checkout to be handed to somebody who is proven to have an interest in the category.
What does that add over simply printing a Product B coupon whenever someone purchases Product A?
Good question and I glossed over that interaction so much that I left out the important differentiator - it can _also_ detect that a product was pulled from the shelf and then put back. That is a much stronger signal of intent, and as a result, that signal can have a price put on it which is higher than just blindly handing out coupons.
In college I needed some cash and went to a “study”. For $75 bucks you went to a mocked up grocery store and spent 45 minutes strapped up in a chair where they showed you combination after combination of (one vendors) bottle colors and arrangement on the shelf and tracked your eyes on how quickly you gravitated towards them. Always seemed super intense what retailers do.
It certainly won't stop them from selling on that feature.
You are right about the lengths product marketing companies will go to. There is a lot of money on the table and the grocery business is all-in on this stuff.
This assumes they will buy multiple of product A? This is largely true for the type of thing Rite Aid sell I'd guess, but perhaps not for higher value items which are less frequently purchased and have higher margins.
It appears that you've been downvoted but I believe you are 100% correct. I lost count of how many times "AI" was mentioned in the one pitch I was in the room for.
Why would you need facial recognition to print a coupon? If the system knows that product A is next to product B, then it can just print B's coupon whenever A is scanned at checkout.
You don't. The system needs to be able to track an individual throughout the store, and can do so via several measures. If you're just tracking a person as a persistent-yet-moving object, there isn't much of a privacy concern. However, if one elects to pile on technologies like facial or gait recognition, then one can track consumer behavior across multiple visits. It's another datapoint which carries it's own value from a marketing perspective.
It just-so-happens to also have value to those charged with loss prevention. If a shoplifter is caught, they can be informed that they are no longer welcome to shop in this store. Enforcing that can be difficult unless....
This sort of thing is a little perverse imo. I don’t think anyone should be punished for stealing food, that’s a crime of hopeless poverty. My roomates were broke in college, spending nearly all money on rent and school. If someone wanted meat that week, they would ring up the chuck as bananas at self checkout. Otherwise it was rice and beans.
OK. What, in your mind, should grocery stores do about shrinkage? Bear in mind that grocery is a notoriously low-margin business, and stores can and will be shut down if shrinkage at a location is too high.
Rice and beans will sustain life. Theft isn't ok. I've used the food bank before and been homeless once, shoplifted once or twice and felt bad about it. Never once was it the last resort, there's always somewhere to get food. As a college student? That's just embarrassing. You can buy 30 packages of ramen for like $2.
Well, anybody starving is immoral. Food theft only hurts profits, not people. Also, ramen is wildly unhealthy being starch and salt with no other nutrients.
"hopeless poverty" != "attending college". If you chose higher education (at some considerable expense if you're in the states) over basic necessities like food, that was your decision. I don't see how it justifies theft.
It sounds more like they made the choice to take out less in loans than they needed than hopeless poverty. I would love for what I'm paying back not to include my food from college, but it does.
Adding to the above, the general goal is to create a physical "heat map" similar to what is used in web eCommerce sites. Fact of the matter, retail is split between a dual role of providing physical locations as well as web locations where they sell their products and services. Because it is easier to base decisions on similar evaluations, the convenience of eCommerce "heat maps" is driving companies to try to create the same for their physical locations. Then all the methods used for surveillance capitalism used online can be applied to their physical retail operations.
Why do people focus so much on how the data can potentially go back to China? It just feels like a redirection to distract us from the privacy violations here in the US. I understand why it's a concern but it just seems like every article always ends up talking mostly about Chinese companies, the Chinese government, how they respond to accusations etc. instead of focusing on our government, our corporations, our security forces all using or wanting to use technology like this.
Take the "Orwellian surveillance" part. We get about 2 paragraphs about how surveillance in the US is largely unregulated and whatnot and then 4 of them about China. The other sections afterwards seem like they were written to almost justify the use of surveillance in US corporations.
Maybe it's all in my head but it's just something I keep noticing whenever there's writing about surveillance in the US.
For the same reason news coverage of things that are bad for everyone focus on how it's bad for black people, or how it's bad for women. Because the press seem categorically incapable of covering negative aspects of society without a uniquely-victimized or uniquely-malicious group involved.
The conventional wisdom is that they need an angle to make readers care. My hot take is that they're too fucking thick to even begin making an intelligent analysis of anything but the most black-and-white topics. If something is actually contentious they just report on a handful of fashionable opinions as co-equal, trashtalk the low-status opinions, and don't even bother acknowledging the principals and/or perceived realities that lead people to various opinions.
I agree with you, however I think a significant portion of the blame for this kind of emergent behavior comes from the consumers of said media. They're producing what they know will sell.
Because it's a national security threat. The real danger of a surveillance state is political/foreign actors using that information to gain advantage. It's one thing for individual's shopping habits to be exposed, it's another thing for the person's information to be used by bad actors for their own purposes.
I suppose this is a controversial opinion then, but I'm much more concerned about my government (US) spying on me than I am on a country I have never stepped foot in (China).
The real danger to me is the US govt based on their track record, and it's acceleration of surveillance and force over the last 4 years
That's fair, but concern about China should not be tossed aside; The U.S. is practically on the verge of a new Cold War with them, where the rules and regulations are nothing like we've ever seen.
> U.S. Senator Marco Rubio, a Florida Republican and acting chair of the U.S. Senate’s intelligence committee, told Reuters in a statement that the Rite Aid system’s potential link to China was “outrageous.” “The Chinese Communist Party’s buildup of its Orwellian surveillance state is alarming, and China’s efforts to export its surveillance state to collect data in America would be an unacceptable, serious threat,” he said.
Paragraph works just as well after s/China/America/g.
If someone s/China/America/g that statement, Mr Rubio would call them an anti-American, far-left communo-fascist who regularly wipes their ass with the constitution, and point out how George Washington personally warned him to keep an eye out for people like that.
I genuinely don't understand what the big deal is. Brick and mortar stores lose a lot lose of money in theft. Walmart alone loses $3 billion each each year. It's their private space, so it makes sense that they want to know who's going inside. The majority of shoppers are using their credit card and store rewards card anyways, so they already know who most of the paying customers are anyways.
For me the issue is that it was apparently done in secret. The difference between CCTV that's stored for a day or two and then permanently deleted and cameras tracking behavior and recording data forever is different enough to require a very different kind of disclosure / signage.
"We track customer movement, behavior and tie that to an individual customer record. We keep this information in perpetuity to serve you better."
Not only that, usually cctvs have a tv facing the entrance with the feed and a reminder that you are on camera. This on the other hand, is Orwellian in comparison.
Would you think it's a big deal if the cops pulled you over, and arrested you for grand theft, took you into booking, finger-printed you, took your picture, etc, then tossed you into a jail cell for a few days because it was a weekend and no judge was available, so you couldn't post bail?
Even after you go through the rigamarole of proving your innocence, you still have to live with having been roughed up by the police, having your mug shoot plastered all over the internet, and having experienced county jail first hand.
And that assumes that you're not charged with something else that happened during the course of the arrest.
The problem is that these systems are deployed in communities such that they implicitly target black people, yet plenty of research shows that they aren't even remotely effective at identifying black people, at all. And the criminal justice system doesn't care.
Given that the legal system targets based on gender more than race, why race is the go to inequality that is pointed out? There seems to be a social level acceptance that unequal targeting based on gender is somehow more acceptable (and by targeting I mean everything from the initial profiling and surveillance all the way to the chance of conviction and the sentence handed out, with each step granting compounding interest on inequality).
The laws on our books might be race-blind, but the original intentions by lawmakers, enforcement by police and prosecutors, and behavior of juries are most definitely not.
“Racism” is one of the most loaded concepts, but it’s a fact that even by this incredibly loaded definition the system is more racist than any other system in all of human history.
But it isn’t racist in the way that you expect it to be. It isn’t racist at the level of the intersection of the system and the “street”.
The problem that you think is the problem, isn’t. The real problem is much more insidious. And if you attempt to solve the problem that you think is the problem you will do nothing but worsen the problem, just as all previous interventions.
Sure, but it's still not the core issue. If racism were magically solved overnight, very little would change. You'd still have these corrupt systems exploiting the weak and poor.
Sort of but not really because it's 1/1000 times the chance that they shop at stores in the same geographic location. So yeah I might have a couple dopplegangers but they're likely spread around the US and can be filtered out relatively easily.
It only becomes a problem if I (1) have a doppleganger, that (2) likes to steal, and (3) lives in the same area as me. It'd still be a kafkaesque nightmare but probably a pretty rare one.
A better way to think of it: if the system only has a 0.1% chance of an incorrect match and at least 700 people regularly shop there, each of them is more likely than not to be misidentified as somebody else.
Think of it more like browser fingerprinting & bits of entropy instead of a one-shot classification. E.g. cross-reference with other data to get near-100% accuracy:" digital fingerprints being emitted from your phone, tracking movement, checking out with a credit card, license plates entering the parking lot, home address vs store location, etc.
If it's not useful, companies aren't going to pay money to use it. As Rite Aid put it, "companies seem to be scaling back or rethinking their efforts around facial recognition given increasing uncertainty around the technology’s utility."
The companies don't care about a few false matches here or there. But if you're one of the people it keeps incorrectly identifying as a criminal who needs to be escorted out of the store, it will matter to you.
Throw in a face mask and this tech is all useless. Especially now that we’ve normalized mask use in the U.S. for probably any cold and flu going forward.
They'll have have access to all your store visits, what your family and friends look like and their store visits, what they buy. And a manager/tech company/security company who may or may not be poor/struggling and VC funding deprived/politically motivated will have at least all this a few clicks away for your lifetime.
Sure they have the right to this, and you can choose whether or not to shop with them if they do, but would you want them to without your consent?
I'm probably OK with it as long as ALL OF the following apply:
(1) Only used within private property. No cameras pointing out into the streets.
(2) Used for security purposes only, not marketing or tracking the customers and selling a profile of them to other organizations.
(3) Adequate IT practices so that access is restricted to only employees who need it and so it isn't hacked into and used for whatever nefarious purposes.
(4) Deployed based on objective criteria based on real data, like loss (theft) rates of stores. Not some regional manager looking at a map and deciding that these stores get it and these others don't based on their subjective idea of what qualifies as a good neighborhood and a bad neighborhood.
(5) Technology can be shown not to treat one group of people worse than another. You don't want scenarios like a system that, when dealing with people of one race, is bad at differentiating individuals, so that a legit ban on a person of that race results in a higher false positive rate for others of that race.
For #2 and #3, it would probably help to just not record (or have very limited retention) of the data.
For #4 and #5, some of this might already be legally required already in the US because of existing civil rights laws that prohibit denying customers access based on race, etc.
E.g. a person could go to a store and either Find or Not Find a camera. For stores where they Find a camera, they can choose to form an opinion about cameras and whatever 'races' shop here, and leave.
E.g. a person could go to a store and see they are on camera. They can make an association about anti-theft cameras if the store is dominantly in a neighborhood of a specific 'race'. Then they go to another neighborhood dominated by a different 'race', from the same Store Chain, and see they are not on camera. They can speculate this neighborhood is trusted more by the store to be lower in theft.
E.g. two people could then be shopping in one store which has no cameras. One person belonging to a 'race' where cameras are typically found is seen by another. The other assumes, based upon prior observation, that this person of that 'race' is associated with theft. They are suspicious of that first person. ==> Bias against a person of a different race.
That's less than 1% of Walmart's total sales. It's not a major loss for them. Especially when Amazon is offering many products at a much cheaper price-point.
I don't support either major retailer, I'm just providing context that within the given sales (and margin), it's barely a drop in the bucket to a problem for Walmart.
It's 0.57% of sales [1]. I tried to find out how much in returns they deal with but...I suspect that returns is a higher number or at minimum comparable.
There was a company called "stop lift" that deployed cameras on its check out lines. Looking for fraud by employees and probably self checkout.
Whenever I'm in a store at the cashiers I look up and often see lots of camera domes above the register.
There's been a trend in self checkout to have at least two cameras directly on each machine, one in the screen pointed at your face, and one overhead on a mast pointed downward.
I wonder if it’s inevitable that we will get denial of entry based facial recognition systems. It’s legally difficult to stop people from leaving the store whereas it’ll be much easier to prevent them from entering.
If someone shoplifts then they get banned from the chain, or a network of chains. This would be an alternative to police involvement. Restitution could be done proportionality and privately.
Of course there will be issues with it but I’m sure it could be cheaper to figure this out than the alternative loss prevention solutions.
From a profit motivation, Rite-Aid (and everyone else) would really like everyone to give them money instead of banning them forever. So I really feel the market will sort this out, maybe a 3 strike system or temporary suspensions.
Now that I think about it, we already have temporary suspensions from Walmart and the like if you get caught. How is this different than a guard just watching the CCTV all day?
It will happen, and it will become a constitutional issue once it becomes ubiquitous enough.
These systems will likely be networked together, building and sharing profiles on known undesirables such that they can be banned from most, if not all stores eventually. Once that happens, there will be a class of people who cannot legally buy food anymore, because to enter a store would be trespass.
A few years back I stayed at the Nugget in downtown vegas. I hate vegas but friends wanted to watch the sweet 16 there. I was actaully somewhat happy with the nugget, it was freshly upgraded and pretty nice. Food was ok and ok priced, and the vegas main street is a better feel for me than the mega hotels.
My friends paid, I paid cash, I skipped checkin and went right to the room.
What flipped was my friend walked up to a blackjack table and the dealer greeted him by name. Another friend who had also checked in was greeted by the cash out teller by name. They didn't seem to know who I was.
I started paying for everything in cash and was like "welp, guess I'm definitely never coming back to vegas"
"the retailer used state-of-the-art facial recognition technology from a company with links to China and its authoritarian government."
And today along with countless millions of North Americans I dialed my mom on a piece of technology that was made in China and its authoritarian Government (TM). FFS can't they stop shoving big bad China into every story. However bad it is not China's fault for US corporation being A-hole.
Added for clarification: This is not to protect China, rather to point out this style of "reporting" that lately seems to have very little to do with the true reporting as it used to be.
After IBM, Microsoft and many others publicly declared (for marketing reasons no doubt) to stop working and delivering face recognition systems, a new competitor quickly fills the space. Color me surprised.
Yes expected. I posted before that by trying to stop companies we work for or other well known companies from operating in this space we give up far too much opportunity to police and possibly influence the outcome.
Facial Recognition is coming along with all other sorts of identification. They key to making sure these are not abuse is to get embedded into the process of making them. This way issues of privacy, accuracy, and accountability, can all be addressed.
Currently far too many here, especially here, are the head in sand type. If they shout it down and declare it evil and see a big name step down they declare it fixed. Ignoring the fact the world is a big place and other companies and countries really don't care what your opinion is. So get in there and make sure where you live that this technology when it does become common place has the structures in place to protect the individual.
Because you can damn well guarantee it won't be corporations abusing it, it will be politically oriented groups who will exploit it. You think the cancel culture is bad now with their name and shame combined with using sycophants to leak records if not outright court challenges to sealed records, wait till they abuse this
"Because you can damn well guarantee it won't be corporations abusing it, it will be politically oriented groups who will exploit it. You think the cancel culture is bad now with their name and shame combined with using sycophants to leak records if not outright court challenges to sealed records, wait till they abuse this
"
I am very worried that there will be a lot going on behind the scenes. Public cancel culture is bad but secret cancel culture is even worse.
> Currently far too many here, especially here, are the head in sand type. If they shout it down and declare it evil and see a big name step down they declare it fixed. Ignoring the fact the world is a big place and other companies and countries really don't care what your opinion is. So get in there and make sure where you live that this technology when it does become common place has the structures in place to protect the individual.
So because it's going to happen anyways, get on board or else?
I'm not a follower of defeatist thinking. Large structural changes can happen, but only through persistence and focus.
Saying "it's not worth it" right from the get-go is a self-fulfilling prophecy.
> They were not responsible for the Nazis, they were only impressed by the Nazi success and unable to pit their own judgment against the verdict of History, as they read it. Without taking into account the almost universal breakdown, not of personal responsibility, but of personal judgment in the early stages of the Nazi regime, it is impossible to understand what actually happened.
Well, sort of. Microsoft and IBM etc. moved out of the space because it was unpalatable to consumers. Now the question is whether the same dynamic will affect Rite Aid, CVS and other major retailers. It's entirely likely that it will, and then you'll see those companies pressured to drop these systems. Articles like this (which bring the practice to public attention -- notice that retailers don't seem eager to disclose it to customers directly) are an important step along the way.
Of course, it's entirely possible that in the end, controlling these practices will require legislation (either to prohibit, or regulate, or at least disclose where this is being done.) If that legislative fight happens, it will be very helpful to not have the most largest and most politically-connected tech firms depending on this stuff as a key revenue stream.
As long as there are no laws there will be companies who will provide these services. There may some companies that still have some ethics and won't serve that market . But the result will that companies with even less ethics will move into that space.
I suspect the companies you are mentioning didn't pull out for ethical reasons but because they have other products and don't need that market.
I think I have to agree with some comments that it's very unlikely companies do get a lot of value from these surveillance programmes. From a side, it all sounds super advanced and promising,but in the reality it's usually more down to earth:
After buying a dehumidifier from Amazon, they keep sending emails with various dehumidifier options.. Like I need 20 of them. Sainsbury's tracks my shopping via Nectar card just to give me 5p discount on the cucumbers and crisps I sometimes buy. Where's all the sophistication? With all the tracking and analytics, one would think they could better track shopper sentiment,yet thousands of tons of unsold food and other products get dumped every day.
Also,just because someone is doing it, that doesn't mean it's profitable. A lot of CEOs get sold all this AI/ML hype and because they have genuine fears of being left out, they just roll with it.
Doesn't seem like a productive battle. It's only upsides for stores that roll out facial recognition. If you want this stopped you will need legislation.
Public shaming didn't work, literally at all. The article was released literally today, and the program was ended before anyone knew about it other than reporters. And this won't deter other corporations from employing this technology.
We absolutely need legislation to permanently stop this. The free market is not going to magically pop up a competing brick and mortar that doesn't spy on us and/or sell/give our data to other corporations/governments.
If everybody rolls it out within a short window of a few years, public shaming will be impossible. If there's no non-participating business that a potential customer can choose instead, shame costs nothing. See Smart TVs.
Facial recognition is used mostly to stop shoplifters so it's to be expected that it will be used in low income neighbourhoods mostly. It is a matter of time until it becomes widespread enough that they won't have to back down for shame.
Do you think it's justified to bring in the use of government force and coercion in a situation where no one is being injured, no one is being defrauded, and people are voluntarily going to the private property on which the events occur?
It would be much more ethical to personally stop going to the store and giving them business if you object (which is a valid opinion, imo). Encouraging others to do the same, spreading news about the practice, etc, all seems good. But once you start calling for people with guns to coerce the property owners you're doing something unethical.
Sounds to me like someone sold Rite Aid a sow's purse, and is no wonder for a chain that had to go begging to its competitors to bail it out of insolvency. (They attempted then failed to merge with CVS and settled for giving more than half of their stores to Walgreens.)
Overcoming masks is a feasible challenge and will be solved with, I assume, modest efforts.
Side note: I tried my Android phone face recognition while using a mask. Interestingly, as long as the nose structure can be perceived, it managed to recognize me.
Acknowledging that you can't change what people will do and that you will be dead soon is the best incognito window I've found.
It's incognito from your stupid brain patterns that want to find something to worry about incessantly. It's not that the stupid thoughts stop - you just pet them like stupid little things that they are and move on with your life.
The article doesn't make clear whether they turned the system off because it wasn't working due to masks, or simply wasn't doing what they meant it to do in the first place.
> “This decision was in part based on a larger industry conversation,” the company told Reuters in a statement, adding that “other large technology companies seem to be scaling back or rethinking their efforts around facial recognition given increasing uncertainty around the technology’s utility.”
Will it come back when the masks go away, or have they realized that it was never useful in the first place?
(They also claim it had nothing to do with race, which is kind of farcical since it's well known these recognition systems have trouble with race. It obviously has something to do with race, even if they're choosing to ignore that issue.)
In brief, my questions are these: Can these systems track the movements of people wearing masks? (I suspect they can, perhaps with some tweaking.) And assuming that these systems can provide accurate movement data through the store, does analysis of that movement data actually provide actionable suggestions of how to arrange the store that, when put to the test, actually increase sales? (I suspect the efficacy of this analysis is greatly overstated by those selling it, and stores are beginning to realize this.)
It's no different than browser fingerprinting. "You can just use a VPN/spoof your user agent!" It's not that simple. Just like how browser fingerprinting has evolved to the point where it's virtually impossible to get around, these real world systems will follow suit. Hair color, eye color, skin color, the types of clothes your wear, your gait, height, gender, path you take through the store, car you drive, license plate, stores you frequent, heck... even where you choose to park or which door you choose to enter could all be used identify individuals/families.
Keep in mind, just like with web analytics, these solutions will be provided by 3rd parties which will aggregate this data across multiple businesses.
It's interesting that it has taken this long to be commercially viable. I have an acquaintance who worked on visual perception and circa 2001 this was a hot research topic for them. IIRC, their team demonstrated a two camera system for gait recognition to LLNL around then.
Next in line: iris scan at entry, followed by tracking you based on clothes/posture. If you don't agree: try to wear and act the same as everone else?
More serious (even though I think my example isn't that far-fetched): I'm really not sure what to think of these camera and whetavere software backend slowly creeping into society. It has been predicted, definitely not everyone is ok with it, but still it just happens.
I was just thinking that...or some type of invisible (to human eye) laser setup. These systems all rely on cameras so why not take advantage of the camera weak spot. Would be awesome to build something that makes you look like a blurred out blob on the camera, but doesn't render their system useless while you're using it.
The quick solution for retailers would be banning use of IR/lasers in-store and they'll play the "if you don't like it you can shop elsewhere" game.
Pill mills were shuttered, but you'd be surprised at how creative people are when they need to feed their habit. There are doctors and pharmacists all over the place that are either:
1.) Being explicitly blackmailed by prescription drug addicts.
2.) Whose employment is completely dependent on prescription drugs because their patients demand prescription drugs as a quick fix for their problems. There are people who literally feed clients to those doctors because they know they are afraid to tell their patients no when they ask for prescription drugs
3.) Don't bother to find alternatives for their patients that don't involve prescription drug abuse because it is a faster method of getting their patients out of the office.
For 1 & 2 specifically;
Doctors receive an itemized DEA report each month for each patient who's prescribed controlled substances. The reports contain all controlled meds prescribed to that patient, from every Dr.
I saw my first DEA report 20-odd years ago, so they're nothing new. Each report is a tap on the shoulder to remind Dr.s that every controlled prescription they write is being closely monitored. Patients are monitored exactly as closely as Dr.s, they just aren't informed every month.
So - the volume of prescriptions inferred by #1 & #2, would immediately attract DEA attention. Both Dr and patient would be in cuffs before drug-seeking patterns had time to develop.
What is entirely more likely than 1&2, is that the endless demonization of prescribing opioids - and the continual LEO threats against Dr's who prescribe meds in a way that the DEA doesn't like - is increasingly frightening Dr.s away from treating pain in an effective way.
Your #3 contains a number of assertions that would be difficult-to-impossible to prove (or disprove), when strung together like that.
I agree with all of your points, including the fact that I do not have supporting data.
Just keep in mind that the people who are really good at skirting prescription drug laws do everything to make their operations look innocuous. They use regular family practices and they dont fill any more controlled substances than any other family practice by percentage of perscriptions. You're never going to see these people showing up in any reports, there won't be any supporting data.
When you cry, “racial discrimination”—“these systems don’t work on black people”, you have implicitly accepted that these systems would be fine if only they worked on black people as well as they work on white people.
I pay with a credit card at every store and pharmacies have access to a ton of my personal health information. These aren't exactly places where you can remain anonymous. I wouldn't have an issue if these systems were near perfect, and had checks-and-balance built in such that false positives don't ruin innocent peoples' lives.
That's my primary issue, is that these point the finger at innocent people, and that innocent person suffers. We have so much evidence of how terribly inaccurate these systems are that it's appalling that they are given any credibility in the legal justice system. That article alone says that their old system misidentified ten people out of ten (!!).
My secondary issue is that these inaccurate systems are profoundly less accurate on PoC, a group of people already subjugated by our criminal justice system.
Taken in isolation, each of these issues should be enough to prohibit their deployment and use.
My suggestion for checks and balances is civil liabilities for those deploying said systems. Since I believe that Rite Aid is going to be pretty fucking diligent if they risk a $100MM lawsuit if they are wrong. I'm of the cynical mindset that the only way to get executives to consider the negative effects of their business strategies is to make it cost money, since that plays to their sociopathic tendency to make everything a dollar-cost optimization.
You can sit there and call me a sucker and a racist for pointing out racial bias in our society. But that's no enough to stop me from standing up for equal rights by calling out racial bias anywhere I see it.
If you want to restore privacy to the people then what you need is a system that is incapable of identifying ten people out of ten, no matter their race.
This is a simple concept.
Yet here you are complaining about the thing that you pretend to be against.
I have no problem with this move. One of my general principles, one I try to apply everywhere, it is: if you can perform an information processing task with your eyes and brain, you should have the freedom to program a computer to do that task. Under this principle, I have to allow public facial recognition: any functioning human can already do it, just at a smaller scale.
On other words, Rite-Aid employees can already recognize your face. Deal with it.
Why is my principle so important? Because the alternative conflicts with the computational enhancement of the mind, which is far more important than any illusory gains to had from banning this tech or that tech.
"Rite-Aid employees can already recognize your face"
Are those same employees networked together and can remember the exact day and time you where in any of the stores and know exactly what you purchased? They have also memorised the entire FBI most wanted list and any other list of faces that may be in a "bad credit" database?
>Under this principle, I have to allow public facial recognition: any functioning human can already do it, just at a smaller scale.
So you'd be fine with mega-corps or the government tracking everyone's movements with drones, and cataloging it into a massive database? After all, you can already do that with a bunch of minimum-wage workers and a map.
>Why is my principle so important? Because the alternative conflicts with the computational enhancement of the mind, which is far more important than any illusory gains to had from banning this tech or that tech.
So basically you want technological singularity, even if it means turning the world into a cyberpunk dystopia?
This of course sounds dramatic. But then, people who fall into this bucket continue to build, sell, and most importantly open source some of the most dramatically societal-changing tech with zero regard for what they enable.
This mindset would have open sourced the Manhattan Project if it was written in Python and used interesting storage advances, but it was safe because it was FOSS.
> Rite-Aid employees can already recognize your face.
You mean the two front store barely-above-minimum wager workers in a 15,000 sq.ft store given 10 tasks to complete in the time available for 6 while still maintaining performance metrics to avoid demerits and who are specifically instructed not to accuse anyone of shoplifting even if they directly observe it happening?
In fact, people can't accurately identity peoples faces, there have been many convictions of people based on people recognizing faces in a line up, and their recall of said persons face being completely wrong.
So by your own logic alone this technology should be dismantled.
You defer to a fantasy utilitarian argument, where the material interests of the average human right now should be subjugated to possible technological advancement. This makes your moral and ethical compass devoid of any humanity. This is truly a horrifying set of morals you have put forth.
Technological progress has always been a net good (life now is better than it has ever been), and luddism (which this facial recognition animus amounts to) has always been a net harm. Welfare has improved massively over time, and delays to technological improvement only delay these welfare gains and add misery to the world.
>Technological progress has always been a net good (life now is better than it has ever been)
Benefits can be imaginary and those don't weigh heavily against the harm being done. Bulk surveillance of people who aren't suspected of a crime, is one example.
Sometimes benefits only outweigh harm because the harm is limited - due to push-back (or the threat thereof) from the public.
It's not just facial recognition in stores: it's tying face, license plate, credit card, and phone app into a single customer view; it's pricing based on your past (and current!) behavior; and it's data consortiums to tie together data across companies (ever wonder why everyone asks for your phone #? It's traditionally the primary key in the big-*ss joins).
The expectation of privacy we have online is coming to the real world, make no mistake. I wish I had a call to action here but... keep yelling, I guess? I don't have a solution.