Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Walmart employees are out to show its anti-shoplifting AI doesn’t work

arstechnica.com

Post reply on HN
Print view View original

81–90 of 102 posts

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#81

Post by Melting_Harps »

logfromblammo wrote: post #64
throwawaygh wrote: post #7

Earlier quoted context omitted.

Right. It takes a lot of theft to pay for a whole extra human standing there. Even a minimum wage.

See also farmers' produce stands that are often completely unmanned and run 100% on the honor system. People take fruit and vegetables, and put the money in the box--or if the farmer is tech-savvy enough, through their phones. And even if a few people don't pay, they still get more money from their inventory than shipping it to a wholesaler or a grocery store. (As far as I know, direct to a local gourmet restaurant p…

> People take fruit and vegetables, and put the money in the box--or if the farmer is tech-savvy enough, through their phones. And even if a few people don't pay, they still get more money from their inventory than shipping it to a wholesaler or a grocery store. (As far as I know, direct to a local gourmet restaurant pays best.)

It does.

> The employee would likely have to stop $200/day of petty theft. That doesn't seem impossible, but the problem is that if you pay your employee minimum wage, they don't actually care that people are stealing from their employer, because they have no actual incentive to do anything beyond what is the bare minimum requirement to avoid getting fired, which may be just catching one shoplifter per day.

> And then, after all, if one caught every shoplifter, every day, people would stop trying, and one might miss quota on later days. The person whose job depends on shoplifters has a perverse incentive to encourage shoplifting, just as cobra-catchers have an incentive to breed and release snakes.

You just explained the the broken window type fallacy of LEO and Police in general. When the crime no longer exists, you create it... by force. Oddly enough, it also applies to loss prevention.

> So the problem is more that it's too expensive to hire someone at a rate high enough to make them care about doing their job well, instead of just keeping it indefinitely while gaming the right metrics.

Agreed. And again, it applies to more than just shoplifting prevention.

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#82

Post by sukilot »

pscoutou wrote: post #75
birdyrooster wrote: post #67

Earlier quoted context omitted.

They must have forgotten the name of the game is loss prevention for the business and not about stopping shoplifting. Cutting payroll on your loss prevention is loss prevention in of itself.

I’m not sure. When self-checkouts were introduced here, they had staff supervising to help and educate customers then significantly scaled back that headcount. Fast forward several years, there’s more staff stationed with self-checkout due to theft. I think retailers are going to calculate the cost of headcount and theft and make a decision based on an equilibrium. Of course, the best outcome for Walmart is to have a…

Self checkout has 1/2 to 1/8 the staff of full checkout.

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#83

Post by logfromblammo »

vonmoltke wrote: post #78
logfromblammo wrote: post #54

Earlier quoted context omitted.

This has already been tried. I, personally, hated it so much that I just stopped using the self-checkouts whenever possible, because employee checkout was faster, even with the wait for customers queued before me. The weight check always brought the checkout process to a screeching halt, requiring an employee to come along and restart it, even with the smallest baskets in the best of conditions.

This was exacerbated if I wanted to use my own bags instead of the provided ones.

Unrecognized item in bagging area!

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#84

Post by sukilot »

alkonaut wrote: post #80
birdyrooster wrote: post #67

Earlier quoted context omitted.

They must have forgotten the name of the game is loss prevention for the business and not about stopping shoplifting. Cutting payroll on your loss prevention is loss prevention in of itself.

Loss is hard to estimate though. What are the losses from a false positive?

One time I had $200 in self checkout and the machine broke and no staff were around so I gave up and walked out. Store lost the sale and whatever goods were considered ruined and I won't go back to that store. but they also saved on staffing costs.

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#85

Post by alkonaut »

sergiosgc wrote: post #37
code4tee wrote: post #16

This is a very hard problem and thinking a product could do this without extremely high error rates is naive. Any implementation should use this as a supplement only, not replacement for humans. For example a human watching 10 checkouts and quickly manually reviewing scenes flagged as suspicious. Any implementation should also account for the fact that loss will still occur and account for that in the RoI. The fact t…

Preventing scanning one item and placing two on the bag? It's an easy problem. Just have scanned items be placed on a scale. People will be bagging stuff to take home, so just have bags placed on a scale and observe the weight increase as items are placed in the bag. You probably know the weight of one jug of milk, but even if you don't, it's easy to "learn" the weights of new SKUs statistically.

I did that weighing in a store and it was an absolute disaster. Cumbersome and error prone, especially when you have larger items etc. I think that has huge hidden losses in people choosing manual checkouts or worse - avoiding the store all together.

Self checkout without that works well in many supermarkets, IKEA etc. There are losses but with random checks they are lower than the staff costs - so It's a win.

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#86

Post by sukilot »

karatestomp wrote: post #41
jjoonathan wrote: post #39

Earlier quoted context omitted.

Their new kiosks are so infernally slow that I like to imagine they're electron apps running on a raspberry pi, which would make them the very peak of innovation.

Same for McDonalds' ordering kiosks. They feel webtechy in the worst way. Plus they're horribly organized, probably to try to boost sales. Unless the line at the counter's over 4 deep I can't imagine the kiosks being faster, even if there's no wait for them.

McDonald's replaced their menus with video ads to pressure customers into spending more and missing the value meals.

So I stopped shopping there. Wendy's FTW.

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#87

Post by sukilot »

karatestomp wrote: post #33
dx87 wrote: post #15

I'm suprised they even bother to try and stop it. My grandfather works at WalMart, and said that they're told to not try and stop shoplifting because they don't want customers to be afraid to shop there.

Some around here hassle people for their receipts near the exit. It was already my least-favorite place to shop for a ton of reasons, so that was enough to stop me ever going. It's one thing at Costco where I agreed to it and they're very fast and nice. I'm not standing in line to get permission to leave friggin' Wal-Mart (yes I know you can just ignore them but they sometimes try to stop you and make a big deal of i…

Are you sure? Receipt checking cannot be forced. I'm surprised corporate would allow that, since they know they can't force the issue so they cannot stop shoplifters.

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#88

Post by logfromblammo »

Melting_Harps wrote: post #81
logfromblammo wrote: post #64

Earlier quoted context omitted.

See also farmers' produce stands that are often completely unmanned and run 100% on the honor system. People take fruit and vegetables, and put the money in the box--or if the farmer is tech-savvy enough, through their phones. And even if a few people don't pay, they still get more money from their inventory than shipping it to a wholesaler or a grocery store. (As far as I know, direct to a local gourmet restaurant p…

> People take fruit and vegetables, and put the money in the box--or if the farmer is tech-savvy enough, through their phones. And even if a few people don't pay, they still get more money from their inventory than shipping it to a wholesaler or a grocery store. (As far as I know, direct to a local gourmet restaurant pays best.) It does. > The employee would likely have to stop $200/day of petty theft. That doesn't s…

For those not familiar with why cobras were mentioned above:

https://en.wikipedia.org/wiki/Cobra_effect

Long story shortened:

  1. British are ruling India
  2. Concern about cobras
  3. Establish dead cobra bounty
  4. Cobra population crashes
  5. Entrepreneurs breed cobras
  6. Bounty program halted
  7. Breeders release stocks
  8. More cobras than before
The source story is just an anecdote, but there are plenty of historically verifiable examples of the effect, not necessarily involving cobras, specifically.

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#89

Post by vmilner »

ggreer wrote: post #6
49531 wrote: post #2

As someone who has inadvertently stolen from Walmart on more than one occasion, I cant attest that their anti-shoplifting AI does not work.

How does one inadvertently steal?

I've done it in the UK (three times) by failing to notice my contactless debit card has not been accepted and walking off with the newspaper/sweets etc. (I've gone back later to give them the money, but this is seems surprisingly difficult for a large store's systems to cope with.)

Re: Walmart employees are out to show its anti-shoplifting AI doesn’t work

#90

Post by pscoutou »

sukilot wrote: post #82
pscoutou wrote: post #75

Earlier quoted context omitted.

I’m not sure. When self-checkouts were introduced here, they had staff supervising to help and educate customers then significantly scaled back that headcount. Fast forward several years, there’s more staff stationed with self-checkout due to theft. I think retailers are going to calculate the cost of headcount and theft and make a decision based on an equilibrium. Of course, the best outcome for Walmart is to have a…

Self checkout has 1/2 to 1/8 the staff of full checkout.

In my example, big grocery chains here had 5-6 employees at launch then dropped it to just one.

Now they have 2-3 employees monitoring the self checkout.

There is a certain point where employers become more focused on reducing shrinkage than headcount.

Post reply on HN