UnitedHealth uses AI model with 90% error rate to deny care, lawsuit alleges

Comments

• bernieecclestoned@sh.itjust.works ⁨1⁩ ⁨year⁩ ago

  Guessing that’s not an error rate, just programmed to refuse first attempts to save cash

  source

  • ellabee@sh.itjust.works ⁨1⁩ ⁨year⁩ ago

    I was in medical billing about 20 years ago, specifically working to get ambulance billing paid by United Healthcare, Blue Cross, whatever. at that time I hated united slightly more than the VA. the VA was a year behind on payment, and they sent a lump check with the list of what it covered separate. but at least they kept track and paid.

    we had to take United Healthcare to the insurance commissioner because their process was deny, then lose the claim, then deny for late billing.

    instead of responding to the insurance commissioner or providing the requested docs or anything, they waited it out, paid the fine, paid the specific claims, and continued as usual.

    so yeah. AI working the way they trained it.

    source
  • EmergMemeHologram@startrek.website ⁨1⁩ ⁨year⁩ ago

    Nonsense.

    It’s most likely just a ~linear regression~ random forest model trained with the loss function of sum((y - y_hat)^2 * cost)

    I have that basic prediction tools are “AI” And they definitely were in 2019.

    source

    • bernieecclestoned@sh.itjust.works ⁨1⁩ ⁨year⁩ ago

      I was being facetious, I have no idea

      source
• ShaunaTheDead@kbin.social ⁨1⁩ ⁨year⁩ ago

  I read about an early study into AI where they were using it to predict whether the pictured animal was a dog or a wolf. It got really good at detecting wolves and when they analyzed how it was determining whether it was a wolf or not, they found that it wasn't looking at the animal at all but instead checking if there was a lot of snow on the ground. If there was, it would say it was a wolf, if there wasn't it would say dog.

  The problem was with the data set used to train the AI. It was doing exactly what it was told. That's the big problem with AI is that it does exactly what we tell it to do, but people are hilariously bad at describing exactly the result they want down to the absolute finest level of detail.

  source

  • Cethin@lemmy.zip ⁨1⁩ ⁨year⁩ ago

    I would describe it more as giving the results we’re asking for rather than doing what we tell it to, but that’s a little too bit of semantics probably. We mostly don’t tell it what to do. We just give it data with some labels and it tries to generate reasons for those labels basically. It’s essentially the issue humans have of “correlation does not equal causation” except with no awareness of this and significantly worse.

    source
• xkforce@lemmy.world ⁨1⁩ ⁨year⁩ ago

  Sigh… and people wonder why I am concerned about AI. It isnt that AI cant be useful, its that I knew right off the bat that it would be applied in the dumbest most haphazard inappropriate ways imaginable.

  source

  • stevedidWHAT@lemmy.world ⁨1⁩ ⁨year⁩ ago

    So why aren’t you concerned with those in power. You chose the low hanging fruit. They do these things with a multitude of technologies and other people not just AI. Just my 2c

    source
  • JCreazy@midwest.social ⁨1⁩ ⁨year⁩ ago

    It’s not the AI I’m concerned about, it’s the companies that are using it without fully understanding how it works and how to implement it properly. They are just throwing it to the wind.

    source
• Jaysyn@kbin.social ⁨1⁩ ⁨year⁩ ago

  I guess we're going to have a decade or so of these companies using AI as a scapegoat before a lawsuit finally makes them responsible for the AI's output.

  source

  • Rentlar@lemmy.ca ⁨1⁩ ⁨year⁩ ago

    Yeah. If there was anything I’d want to come from this lawsuit if not the company effectively being tried for murder, it would be the need for insurance companies to disclose the data and reasons behind denials which they claim are proprietary.

    source
• Diplomjodler@feddit.de ⁨1⁩ ⁨year⁩ ago

  I can write a script like that in Python in five minutes. Just send me a check.

  source

  • Chariotwheel@kbin.social ⁨1⁩ ⁨year⁩ ago

    I can do it in one minute with PHP.

    <?php
    
    denyClaim();
    
    ?>
    

    That might actually have a better change to be correct that the UnitedHealth AI,

    source

    • ThePantser@lemmy.world ⁨1⁩ ⁨year⁩ ago

      I could flip a coin and have a lower error rate.

      source
• stevedidWHAT@lemmy.world ⁨1⁩ ⁨year⁩ ago

  This stuff is gonna go either of a few ways:

  • govt takes side of ignorant people and restricts the use of AI. Now all the rich and elite have sole access to one of many kinds strongest and most promising creations. (Among many I’m sure)

  • govt makes ineffective policy to restrict companies, which Republicans will then loosen and they’ll spiral back and forth for who knows how long

  • aliens

  • Covid XII the 14th^2 electric boogaloo

  • AI becomes sentient in some dudes basement and starts replicating over the internet to open devices and makes a case for itself better than anyone ever has either making the religious fanatics spaz out (despite god giving the power to control breeding since the get go of both man and animal I.e shepherds)

  What are your scenarios you see playing out?

  source