This is truly sad. If this technology is going to work like claimed in the future and diagnosed early breast cancer in just 5% of people it could help save 32000 people a year and help a lot of people who don't die but have to go though a horrible treatment process.
He'll spent billions to treat himself and will still do this kind of shit.
Always remember these people consider themselves above the crowd and think it's completely normal they're treated differently.
My understanding is that this isn't what happened.
Meta laid off a bunch of employees. Meta claims they used an AI to gauge employees' productivity as a basis for who they laid off. The employees are accusing Meta of laying off employees who took too much medical leave.
Lives that can be saved via early treatments, instead used to profile people based on their health conditions, disabilities and other factors ismorbidly sickening,
We truly do think that Neural Learning Model can be a strong tool as a data sorting algorithm in health and science, but cannot give us answers itself, only point out things we can’t see
It can happen to every and any healthcare system, insurance, institution having data in the cloud: if the host has activities in the US, it's at legal risk of data being demanded under US laws ; if not, there's always the risk of a leak.
That Meta used it so directly is very Meta. But you don't know what your employer has on you thanks to big data. You're in debt? Well, that means you desperately needs the job: no bonus, no raise. You have health issue? Let's put you on the red list: no promotion, favored name in case of layoff. Etc.
But when I say 'skills' I didn't mean individual coding ability, I mean like Analytics. Skills that an entire corporation wields. Monitoring weather and hiking prices for utilities during a national emergency incrementally or 'intelligently'. The way VW cheated on emissions. The way flock uses optical text recognition to read license plates and effectively creates a massive surveillance apparatus.
Use tech to do a thing, but scale it up wide and quick, abuse it before the law can catch up.
The Supreme Court just ruled that it's not racism unless someone specifically points at an action and says "I'd like to do that racism, please!" and so I expect them to use that precedent to target people with disabilities. "They didn't say it out loud, and even though a policy specifically harmed the disabled, it doesn't count"
Ah you see your honour it can't be illegal, because it's actually more profitable this way, and so conform to the natural order of the brotherhood of the ever rising line (forever may it rise, straight and true)
This is basically it and how I see it working at places I've worked at. HR gathers information about you and then uses it to fire you when firing you is necessary....when you're no longer needed. So for example if you're an assembly worker and your product is end of production, you're done. Now engineering goes to work on a new trinket and you gotta wait until they are done. But when engineering is done, they're fucked. Now you're king because there's production to be done while engineering scrambles to figure out a new project that management wants to work on.
Alright, if you look at this story from a more reputable source (Reuters), it's quite a bit different. Meta isn't accused of using AI with people's medical information to make early diagnoses and lay them off. Instead,
According to the complaint, Meta used a number of internal AI-assisted systems to score and rank employees on a termination list. Those included "Metamate," a large language model assistant; an employee-trained "second brain" that tracked workers' communications and documents; and a productivity score drawn from scanning keystrokes, screen content, emails and browser history, according to the lawsuit.
The 26 plaintiffs, who filed the lawsuit anonymously, are accusing Meta of violating federal and state laws that ban discrimination or retaliation against workers who have disabilities, take medical leave or are pregnant. They also claim that Meta failed to test its AI systems for bias in violation of recently adopted California and New York City laws.
So it's still shitty workplace surveillance and discrimination but not the same kind of surveillance and discrimination alleged in the Twitter posts. Meta is using AI to try to rank workers' productivity (on a bunch of shitty metrics) in a way that may discriminate against people with disabilities or who take medical leave or who are pregnant.
Decl. ¶¶ 19; Doe 4 Decl. ¶¶ 11. The monitoring program—announced internally as the
“Model Capability Initiative”—thus captured not only employees’ work output but also
their protected-leave and medical activity.
I literally said that Meta was firing people for protected leave, disabilities, and medical issues. What I was denying is that they were firing people on the basis of "pre-diagnoses." All the complaints in the lawsuit are based on diagnoses that employees disclosed to Meta. I think that's shitty and illegal, but I don't think there is evidence that Meta is acquiring undisclosed medical information about their employees and using that for layoffs.
Reuters is not a reputable source. And it's the same headline
Meta isn’t accused of using AI with people’s medical information to make early diagnoses and lay them off. Instead,
Because Meta doesn't have access to their medical info yet. They would 100% do so if they ever managed to get their hands on this info. Big corporations can't wait to go full Aktion T4 mode.
No idea why they are saying it’s not reputable. As far as I am aware, it puts out probably some of the most trusted news compared other similar news outlets. Plus it is also one of the least biased news sources out there right now.
Oh! Looks like there is an interactive version now. It does not seem to be optimized for phone, but Reuters is still roughly in the same spot it was in 2018.
You're just posting the Propaganda Multiplier but it's the propaganda version of it. These newspapers get their talking points from the US government and will publish whatever they are asked to. This is especially noticeable whenever the US invades a country and lies to manufacture consent for the invasion.
You're posting a picture of Swiss Propaganda Research, an anonymous group that spreads conspiracy theories while accusing Swiss media of "one-sided and uncritical" reporting. This falls squarely into what we in Switzerland call Schwurbler territory: people who peddle wild claims like Bill Gates implanting 5G chips through COVID vaccines or "mainstream science" secretly working to enslave humanity. The figure presumably behind SPR, Daniele Ganser, styles himself as a "peace researcher" (a self-invented title with no academic foundation). He was fired from ETH Zurich, where he served as a Senior Researcher, for promoting 9/11 conspiracy myths, and his subsequent teaching positions, including at the University of St. Gallen, were terminated after his theories were deemed damaging to institutional reputation.
It doesn't matter who made it. What matter is that it's correct. All international media copy pastes anything these propaganda outlets say without double checking it.
If you want a more blatant example of Reuters lying, during the Maccabi riots in Amsterdam they said that a video of Maccabi supporters attacking a Moroccan was the inverse to push the narrative that there was a Pogrom against Jews
Reuters and other propaganda outlets were notified of this by the person who made the video but they refused to correct it for multiple days because they had to keep the narrative going.
I will not engage further in this convo. you're simplifying things. nobody claims that media never is without a doubt biasfree. but stating that all is propaganda fed by the pentagon is a false claim and spreading such nonsense is dangerous for every democracy. tschüss.
Yeah..... I've kinda been dreading the day when they attempt to modify HIPAA to allow the automation of the authorization process for work places to access your medical data.
Right now as a provider if a workplace requests medical records for one of my patients I get to decide if they have authorization, a valid reason to make the request, and get to determine what information they have access to that is pertinent to their request.
I have a sneaking suspicion that with the spread of EMR systems like Epic, employers will attempt to pressure legislators to change HIPAA to automate the process and cut providers out of the picture.
The problem is that electronic medical records programs like Epic are still really shit about how they allocate things like diagnosis codes. There doesn't seem to be a good way to differentiate primary diagnosis codes made by a PCP and the diagnosis codes used for billing and documenting follow up care. It forces you to enter diagnosis codes anytime you enter a medical note, even if you don't have the proper licensing to make a diagnosis. So when someone like a nurse is documenting their care, they may just add miscellaneous diagnosis code just so they can enter their documentation.
Now any human who works in healthcare and has used Epic can read through the note and from context understand that it's a throw away miscellaneous dx code, but something like AI may just use that dx code as justification to widen the scope of their access.
Probably any system that is managed too efficiently starts to suck.
This is really an interesting example. Not sure if one should blame AI or capitalism for this. Kinda reminds me of Gattaca.
Theoretically capitalism could do this manually to optimize for profit, but there are limits. You can only hire so many completely psychotic human resources employees until very weird and bad shit starts to happen and your company is sucked straight into a hellmouth. But an AI robot is completely objective, unemotional, unbiased, beautifully soulless and can surely maximize profit endlessly with no pesky human side effects. Obviously it can't but that is the dream.
Wow this is a whole another level of low for America. Still this doesn’t reach to their permanently imprinted slavery history but they’re regressing rapidly in that level.
Any other country would do this? I really don’t think so.
Funny that this isnt even "AI".
This is image and pattern recognition software that's been used in medicine for years and years.
The AI part is that it's scraping all your medical data together into a single RAG so they can use it to fuck you over as efficiently as possible in the name of making more money.
Pattern recognition was “AI” until LLM took over almost all meaning of the phrase. One of the reasons corporations were supposed to store everything (storage is cheap) was to feed pattern recognition models.
That's the problem with using pseudo-science and bullshit terminology to get cash. They constantly need to increase their bullshit to get more funding.
machine learning actually seems pretty well defined, standard across industry and academia, and we don't have any better words to describe the category
64 replies
Oh look, psychopath CEOs are helping people die faster. And we still do nothing.
This is truly sad. If this technology is going to work like claimed in the future and diagnosed early breast cancer in just 5% of people it could help save 32000 people a year and help a lot of people who don't die but have to go though a horrible treatment process.
Sounds like zuck needs to have irreversible health issues~
He'll spent billions to treat himself and will still do this kind of shit.
Always remember these people consider themselves above the crowd and think it's completely normal they're treated differently.
That's not what they mean.
The better question is why does their employer have access to their private medical data
My understanding is that this isn't what happened.
Meta laid off a bunch of employees. Meta claims they used an AI to gauge employees' productivity as a basis for who they laid off. The employees are accusing Meta of laying off employees who took too much medical leave.
This reminds me of the movie Gattaca. A future world where your path in life is heavily influenced based on your genes.
Unlike our noble society of today where people are not judged based on looks, intelligence, strength or people skills.
Lives that can be saved via early treatments, instead used to profile people based on their health conditions, disabilities and other factors ismorbidly sickening, We truly do think that Neural Learning Model can be a strong tool as a data sorting algorithm in health and science, but cannot give us answers itself, only point out things we can’t see
This sounds American.
It can happen to every and any healthcare system, insurance, institution having data in the cloud: if the host has activities in the US, it's at legal risk of data being demanded under US laws ; if not, there's always the risk of a leak.
That Meta used it so directly is very Meta. But you don't know what your employer has on you thanks to big data. You're in debt? Well, that means you desperately needs the job: no bonus, no raise. You have health issue? Let's put you on the red list: no promotion, favored name in case of layoff. Etc.
Fuck yeah
🇺🇲🇺🇲eagle screech that's actually a red tailed hawk 🇺🇲🇺🇲
"Technology" under capitalism is a regressive force for violence, theft, and control.
Technology enabled new skills, and then these are used with force to gain power.
Technology enabled current skills to be done at a larger scale, essentially creating new skills that are then applied with force to gain power.
The people with the skills to use technology are not the ones with the power.
Their bosses are the ones with power, and they don't know fucking anything.
Yup, exploitation of labor per usual.
But when I say 'skills' I didn't mean individual coding ability, I mean like Analytics. Skills that an entire corporation wields. Monitoring weather and hiking prices for utilities during a national emergency incrementally or 'intelligently'. The way VW cheated on emissions. The way flock uses optical text recognition to read license plates and effectively creates a massive surveillance apparatus.
Use tech to do a thing, but scale it up wide and quick, abuse it before the law can catch up.
The Supreme Court just ruled that it's not racism unless someone specifically points at an action and says "I'd like to do that racism, please!" and so I expect them to use that precedent to target people with disabilities. "They didn't say it out loud, and even though a policy specifically harmed the disabled, it doesn't count"
Edit. Louisiana v Callais
https://en.wikipedia.org/wiki/Louisiana_v._Callais
Ah you see your honour it can't be illegal, because it's actually more profitable this way, and so conform to the natural order of the brotherhood of the ever rising line (forever may it rise, straight and true)
The SC just went to Congress yesterday to beg for money for security. I wonder why they need security if they are doing their jobs?......
This is basically it and how I see it working at places I've worked at. HR gathers information about you and then uses it to fire you when firing you is necessary....when you're no longer needed. So for example if you're an assembly worker and your product is end of production, you're done. Now engineering goes to work on a new trinket and you gotta wait until they are done. But when engineering is done, they're fucked. Now you're king because there's production to be done while engineering scrambles to figure out a new project that management wants to work on.
Alright, if you look at this story from a more reputable source (Reuters), it's quite a bit different. Meta isn't accused of using AI with people's medical information to make early diagnoses and lay them off. Instead,
So it's still shitty workplace surveillance and discrimination but not the same kind of surveillance and discrimination alleged in the Twitter posts. Meta is using AI to try to rank workers' productivity (on a bunch of shitty metrics) in a way that may discriminate against people with disabilities or who take medical leave or who are pregnant.
What a bullshit arguments.
Literally in the court case papers.
source: https://storage.courtlistener.com/recap/gov.uscourts.cand.474171/gov.uscourts.cand.474171.1.0.pdf
You will now say this is not true ? You're AI enthusiast, Meta employee or something ?
I literally said that Meta was firing people for protected leave, disabilities, and medical issues. What I was denying is that they were firing people on the basis of "pre-diagnoses." All the complaints in the lawsuit are based on diagnoses that employees disclosed to Meta. I think that's shitty and illegal, but I don't think there is evidence that Meta is acquiring undisclosed medical information about their employees and using that for layoffs.
Reuters is not a reputable source. And it's the same headline
Because Meta doesn't have access to their medical info yet. They would 100% do so if they ever managed to get their hands on this info. Big corporations can't wait to go full Aktion T4 mode.
I always heard Reuters is very reputable. Can you share some evidence that shows why you think it isn't?
No idea why they are saying it’s not reputable. As far as I am aware, it puts out probably some of the most trusted news compared other similar news outlets. Plus it is also one of the least biased news sources out there right now.
From https://guides.emich.edu/WRTG121/evaluating
Is there a newer version of this chart? It's dated from 2018, and things have certainly changed since then.
Oh! Looks like there is an interactive version now. It does not seem to be optimized for phone, but Reuters is still roughly in the same spot it was in 2018.
https://app.adfontesmedia.com/chart/interactive
One Palestinian genocide whitewashing later and Reuters is still on top...
Question the raters for once.
As a non American from a more socialist country: more like Overtonwindowbias.com 😅
Thanks, that's what I understood!
You're just posting the Propaganda Multiplier but it's the propaganda version of it. These newspapers get their talking points from the US government and will publish whatever they are asked to. This is especially noticeable whenever the US invades a country and lies to manufacture consent for the invasion.
You're posting a picture of Swiss Propaganda Research, an anonymous group that spreads conspiracy theories while accusing Swiss media of "one-sided and uncritical" reporting. This falls squarely into what we in Switzerland call Schwurbler territory: people who peddle wild claims like Bill Gates implanting 5G chips through COVID vaccines or "mainstream science" secretly working to enslave humanity. The figure presumably behind SPR, Daniele Ganser, styles himself as a "peace researcher" (a self-invented title with no academic foundation). He was fired from ETH Zurich, where he served as a Senior Researcher, for promoting 9/11 conspiracy myths, and his subsequent teaching positions, including at the University of St. Gallen, were terminated after his theories were deemed damaging to institutional reputation.
https://www.beobachter.ch/multimedia/digital/anonyme-warner-mit-scharfem-s-297308
It doesn't matter who made it. What matter is that it's correct. All international media copy pastes anything these propaganda outlets say without double checking it.
If you want a more blatant example of Reuters lying, during the Maccabi riots in Amsterdam they said that a video of Maccabi supporters attacking a Moroccan was the inverse to push the narrative that there was a Pogrom against Jews
Reuters and other propaganda outlets were notified of this by the person who made the video but they refused to correct it for multiple days because they had to keep the narrative going.
https://whispering.media/the-maccabi-gospel/
I will not engage further in this convo. you're simplifying things. nobody claims that media never is without a doubt biasfree. but stating that all is propaganda fed by the pentagon is a false claim and spreading such nonsense is dangerous for every democracy. tschüss.
They’re on lemmy.ml. I’m sure they have some insane tankie reasoning
Your server is banning people for celebrating the death of Lindsey Graham it's kind of incredible you're claiming people are insane to not be on it.
Here's a recent flagrant one https://lemmy.ml/post/48321919
Yeah..... I've kinda been dreading the day when they attempt to modify HIPAA to allow the automation of the authorization process for work places to access your medical data.
Right now as a provider if a workplace requests medical records for one of my patients I get to decide if they have authorization, a valid reason to make the request, and get to determine what information they have access to that is pertinent to their request.
I have a sneaking suspicion that with the spread of EMR systems like Epic, employers will attempt to pressure legislators to change HIPAA to automate the process and cut providers out of the picture.
The problem is that electronic medical records programs like Epic are still really shit about how they allocate things like diagnosis codes. There doesn't seem to be a good way to differentiate primary diagnosis codes made by a PCP and the diagnosis codes used for billing and documenting follow up care. It forces you to enter diagnosis codes anytime you enter a medical note, even if you don't have the proper licensing to make a diagnosis. So when someone like a nurse is documenting their care, they may just add miscellaneous diagnosis code just so they can enter their documentation.
Now any human who works in healthcare and has used Epic can read through the note and from context understand that it's a throw away miscellaneous dx code, but something like AI may just use that dx code as justification to widen the scope of their access.
So did they look at weird boobie pictures to decide to fire female employees or nah?
They would if they could. That's the point.
They're already using AI to detect medical conditions to fire people on.
Capitalism will eventually look at your weird boobie pictures and fire you for them if you Capitalism long enough.
Probably any system that is managed too efficiently starts to suck.
This is really an interesting example. Not sure if one should blame AI or capitalism for this. Kinda reminds me of Gattaca.
Theoretically capitalism could do this manually to optimize for profit, but there are limits. You can only hire so many completely psychotic human resources employees until very weird and bad shit starts to happen and your company is sucked straight into a hellmouth. But an AI robot is completely objective, unemotional, unbiased, beautifully soulless and can surely maximize profit endlessly with no pesky human side effects. Obviously it can't but that is the dream.
Capitalism exists to serve, you guessed it, capital.
This means those who have money will be served and those who don't have money will not.
Big banks who have money are getting bailouts and poor people have their social security removed.
Large companies have their needs fulfilled and those who don't have money are not.
Only in 'Murica would life-saving developments be used to deny people treatment so that some rich cunts can accumulate even more wealth
It's a broken country, and I can't wait to see it fall
Broken implies that this wasn’t the goal.
the goal is broken
Anything that can be exploited for profit is fair game here. There's hardly anything left at this point, so it's starting to get grizzly
Reason 748 on why the US health system is fucked up.
Anywhere else in the world, this saves lives.
Reason 748 on why the US is fucked up.
FTFY
Wow this is a whole another level of low for America. Still this doesn’t reach to their permanently imprinted slavery history but they’re regressing rapidly in that level.
Any other country would do this? I really don’t think so.
Funny that this isnt even "AI". This is image and pattern recognition software that's been used in medicine for years and years. The AI part is that it's scraping all your medical data together into a single RAG so they can use it to fuck you over as efficiently as possible in the name of making more money.
Pattern recognition was “AI” until LLM took over almost all meaning of the phrase. One of the reasons corporations were supposed to store everything (storage is cheap) was to feed pattern recognition models.
That's the problem with using pseudo-science and bullshit terminology to get cash. They constantly need to increase their bullshit to get more funding.
TBH that's not "AI" either. In reality there's no such thing as "AI" at all. 100% grift.
In general people don't really hate AI itself but the absolute shitcocks in charge of it and their methods of using it.
It's still machine learning
No, it's not "machine learning" either. That's just another meaningless grifter term.
machine learning actually seems pretty well defined, standard across industry and academia, and we don't have any better words to describe the category
The personification of all evil, the most impossibly evil creature, would come to earth just to learn from our rich how to properly do evil.
When was this posted? There’s no context in the image.
19 hours ago