4 ms·

There is a lot of information in the article but little to draw a conclusion from. They bring up examples where the AI failed, but that alone does not indicate

There is a lot of information in the article but little to draw a conclusion from. They bring up examples where the AI failed, but that alone does not indicate it doesn't work. An AI that fails to recognize some thefts but does recognize others can still be considered to work as long as enough thefts are caught. Imperfect, but given the low cost of a failure relative to not using an AI, imperfection seems quite tolerable. Even the false positives seem to be of little concern as they involve a shopping associate coming over and having to assist an individual. As long as the experience for the shopper remains more pleasant than waiting in line for one of the cashier operated checkouts, it is still an overall win. So yes, the AI is imperfect, but where is the information that justifies the conclusion it doesn't work and not the conclusion that it doesn't work perfectly?

microcolonel 6y ago
Probably the reasons for attempting to undermine it are more political than practical or ethical.
jariel 6y ago
It may very well be true, and in fact reasonable that there's a lot of false positives and that 'it was designed that way' due to obviously an inherently higher error rate. So it's really a that the parameters of the tech are overlapping poorly with the parameters of customer convenience, perception, treatment, and other aspects of the next consumer experience and op. costs. This is probably a great example of understanding the limitations and special needs of AI in the real world. It's not magic, it's just 'some magic' that we have to develop processes to work with.
wyck 6y ago
It is a bit strange how the in-house and other competitor is also mentioned, Walmart are big on data warehousing and management, in a lot of cases they are pushing innovation and r&d, in some cases for decades. This article needs to show a more detailed analysis of when it does actually work, a fraud KPI ?, or something else besides some anonymous complaints. You can trick any system, but that's not telling on it's overall effectiveness.
- voz_ 6y ago
  Walmart pushing innovation? Citation needed.
  - jjoonathan 6y ago
    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.
    - karatestomp 6y ago
      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.
      - sukilot 6y ago
        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.
    - luckylion 6y ago
      Or they have "time spent in store" as a KPI because the more people are in the store, the more they'll buy... and slow kiosks will keep you around.
    - acheron9383 6y ago
      You may not like it, but this is what peak tech looks like.
  - m0zg 6y ago
    Walmart has been pushing innovation for decades now: https://www.walmartlabs.com/ https://www.walmartlabs.com/. How do you think they got so huge?
  - 6y ago
birdyrooster 6y ago
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.
- pscoutou 6y ago
  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 an automated LP system to reduce staffing cost for the long term.
  - sukilot 6y ago
    Self checkout has 1/2 to 1/8 the staff of full checkout.
    - pscoutou 6y ago
      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.
      - throwawaygh 6y ago
        If they’re open 24/7/365, that’s still six figures in savings. And if one of those employees is watching the self checkouts then losses might not even be that much higher. And that with 90s technology — very little/no CV. I think Cashier as a stand-alone job will probably die off in our lifetime. Maybe you still have a store manager who is up at the front of the store whenever they are not doing something else. (I’ve been a cashier at a grocery store)
  - 6y ago
Cd00d 6y ago
I think what people fail to recognize is that AI/ML solutions will _always_ be imperfect. But, they scale. There's always a threshold for the accuracy/efficiency point. It's not meaningful to identify errors, even if they seem glaring, if overall you're reducing 80% of loss at 10% of the expense.
- yencabulator 6y ago
  Human solutions are _also_ always imperfect. On top of that, you have to also account for the concept that the employees themselves might steal from the store.
  - Cd00d 6y ago
    Agreed. Which is actually an upside to an automated, but imperfect solution as reinforcement. I've just been generally surprised at how often I see an ML solution implemented the complaint is "it isn't perfect", when perfection was not the goal metric. I think people just expect robots to do everything exactly right.

microcolonel 6y ago

Probably the reasons for attempting to undermine it are more political than practical or ethical.

jariel 6y ago

It may very well be true, and in fact reasonable that there's a lot of false positives and that 'it was designed that way' due to obviously an inherently higher error rate. So it's really a that the parameters of the tech are overlapping poorly with the parameters of customer convenience, perception, treatment, and other aspects of the next consumer experience and op. costs. This is probably a great example of understanding the limitations and special needs of AI in the real world. It's not magic, it's just 'some magic' that we have to develop processes to work with.

wyck 6y ago

It is a bit strange how the in-house and other competitor is also mentioned, Walmart are big on data warehousing and management, in a lot of cases they are pushing innovation and r&d, in some cases for decades. This article needs to show a more detailed analysis of when it does actually work, a fraud KPI ?, or something else besides some anonymous complaints. You can trick any system, but that's not telling on it's overall effectiveness.

voz_ 6y ago
Walmart pushing innovation? Citation needed.
- jjoonathan 6y ago
  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.
  - karatestomp 6y ago
    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.
    - sukilot 6y ago
      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.
  - luckylion 6y ago
    Or they have "time spent in store" as a KPI because the more people are in the store, the more they'll buy... and slow kiosks will keep you around.
  - acheron9383 6y ago
    You may not like it, but this is what peak tech looks like.
- m0zg 6y ago
  Walmart has been pushing innovation for decades now: https://www.walmartlabs.com/ https://www.walmartlabs.com/. How do you think they got so huge?
- 6y ago

birdyrooster 6y ago

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.

pscoutou 6y ago
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 an automated LP system to reduce staffing cost for the long term.
- sukilot 6y ago
  Self checkout has 1/2 to 1/8 the staff of full checkout.
  - pscoutou 6y ago
    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.
    - throwawaygh 6y ago
      If they’re open 24/7/365, that’s still six figures in savings. And if one of those employees is watching the self checkouts then losses might not even be that much higher. And that with 90s technology — very little/no CV. I think Cashier as a stand-alone job will probably die off in our lifetime. Maybe you still have a store manager who is up at the front of the store whenever they are not doing something else. (I’ve been a cashier at a grocery store)
- 6y ago

Cd00d 6y ago

I think what people fail to recognize is that AI/ML solutions will _always_ be imperfect. But, they scale. There's always a threshold for the accuracy/efficiency point. It's not meaningful to identify errors, even if they seem glaring, if overall you're reducing 80% of loss at 10% of the expense.

yencabulator 6y ago
Human solutions are _also_ always imperfect. On top of that, you have to also account for the concept that the employees themselves might steal from the store.
- Cd00d 6y ago
  Agreed. Which is actually an upside to an automated, but imperfect solution as reinforcement. I've just been generally surprised at how often I see an ML solution implemented the complaint is "it isn't perfect", when perfection was not the goal metric. I think people just expect robots to do everything exactly right.