this post was submitted on 21 Jan 2026
275 points (98.9% liked)

Fuck AI

5268 readers
2307 users here now

"We did it, Patrick! We made a technological breakthrough!"

A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.

AI, in this case, refers to LLMs, GPT technology, and anything listed as "AI" meant to increase market valuations.

founded 2 years ago
MODERATORS
 

In a conversation at this year's rich person convention—aka the World Economic Forum—Microsoft CEO Satya Nadella warned that AI will lose public support unless it's used to "do something useful that changes the outcomes of people and communities and countries and industries."

He did at least provide one real example of what he means by all this: "When a doctor can … spend more time with the patient, because the AI is doing the transcription and entering the records in the EMR system, entering the right billing code so that the healthcare industry is better served across the payer, the provider, and the patient, ultimately—that's an outcome that I think all of us can benefit from."

top 50 comments
sorted by: hot top controversial new old
[–] buddascrayon@lemmy.world 9 points 14 hours ago

He did at least provide one real example of what he means by all this: "When a doctor can … spend more time with the patient, because the AI is doing the transcription and entering the records in the EMR system, entering the right billing code so that the healthcare industry is better served across the payer, the provider, and the patient, ultimately—that's an outcome that I think all of us can benefit from."

That shit can't even set a timer for my eggs correctly. Have a fuck is it supposed to do all that shit?

You never had the “social permission” to do any of this idiotic bullshit, you absolute fucking wilted parsnip.

[–] bytesonbike@discuss.online 81 points 23 hours ago (2 children)

The Internet calling it MicroSlop is probably getting to him.

From tech bloggers calling them out in adding AI to everything including Notepad, to journalists raising eyebrows to their confusion why other companies aren't shoving AI into everything, to this:

https://lemmy.nz/post/33323830

[–] Akh@lemmy.world 19 points 22 hours ago (1 children)

This is amazing… hundreds of billions poured i to this shitshow whereas if they just dumped hundreds of billions into housing, healthcare, schools, that would have infinitely better outcomes for the public

[–] tomiant@piefed.social 4 points 12 hours ago (1 children)

That would mean sharing. Which is communism. Which is evil and leads to people not having housing, healthcare, and schools.

[–] frosty@pawb.social 2 points 2 hours ago

And less money for the CEO. Think of the yachts and leather jackets they could have had!

[–] KindnessIsPunk@lemmy.ca 15 points 21 hours ago (1 children)

I was a die hard windows user that recently installed Linux and suffered through the learning curve because windows was getting that frustrating

[–] bizzle@lemmy.world 3 points 17 hours ago (1 children)

How are you liking it? Was the transition as hard as you thought?

[–] KindnessIsPunk@lemmy.ca 1 points 16 hours ago

There have been speed bumps with things not working intuitively or strange bugs or being esoteric issues with server hosting requiring significant time spent googling but after overcoming the initial learning curve it has been pleasant but I must admit I'm still early on my journey.

Discord works this time, streaming and all which is an improvement from last time.

[–] wonderingwanderer@sopuli.xyz 30 points 20 hours ago (1 children)

They already don't have "social permission" to do what they're doing, so what are they worried about losing?

Also, an american tech oligarch speaking to a forum in Europe about billing healthcare insurance companies is fucking hilarious.

Everyone else was probably thinking "Does he realize we're not all american?"

[–] antonim@lemmy.dbzer0.com 5 points 20 hours ago (1 children)

It's a forum in Europe but many businessmen there were certainly from the US. It's the World Economic Forum, after all.

[–] wonderingwanderer@sopuli.xyz 5 points 20 hours ago* (last edited 20 hours ago)

Yes, but not exclusively from the US. It's the World Economic Forum, after all.

Maybe the other american oligarchs were all "Ta-ta," but that doesn't make it any less ignorant to resort to american defaultism before a worldwide forum...

[–] MountingSuspicion@reddthat.com 58 points 23 hours ago (5 children)

I do not want AI involved in my patient-doctor communication at all. If transcription software is needed, though I'm not convinced it is, then they can use transcription software, but at the end of the day I think a human being should be the one responsible and making decisions regarding what is and is not officially listed in a medical record. AI is not sufficiently advanced enough for me to trust that it will not make mistakes that could endanger lives.

If we wanted to save time with billing codes, we could just do away with them and have a system that just lets people get the healthcare they need. If a test is ordered, that test should be entered as is by the doctor and not need any additional interpretation or overhead. I don't do medical billing, but I can't imagine a reason it needs to be more complicated than that.

Specialized AI double checking radiology may have a use, but I still don't see it as a replacement as much as a second check.

[–] tomiant@piefed.social 2 points 12 hours ago

“A computer can never be held accountable. Therefore a computer must never make a management decision.”

[–] Denjin@feddit.uk 27 points 22 hours ago (1 children)

Specialized AI double checking radiology may have a use, but I still don't see it as a replacement as much as a second check.

Exactly. An automatic second opinion that flags scans that should be looked at again? Excellent. Something that's wrong 10% of the time being the sole decision maker? No thank you.

[–] Valthorn@feddit.nu 7 points 21 hours ago

Remember a few years ago when articles about neural networks scanning for cancer found stuff before human doctors did? That's a form of AI I'm afraid will get defunded too when that hammer comes down on all this "LLM in everything" bullshit.

[–] Get_Off_My_WLAN@fedia.io 9 points 21 hours ago (1 children)

Even the transcription software that is not purely transcription, but actually an LLM that "takes notes" and "summarizes" can hallucinate and add in stuff that was never mentioned. And then AI trained on biased data (i.e., like all of them) tend to give shittier notes for any patients who are of color.

[–] MountingSuspicion@reddthat.com 2 points 17 hours ago

Oh definitely. That's exactly why I specified that it needs to be explicitly transcription software. Even speech to text on my phone gets it wrong with enough regularity that I check it every time. I can't imagine what it would be like if I was using less common words In a medical setting. I don't love the idea of every word said in a doctor's office being recorded and that recording being on record forever, but to a certain extent I can understand doctors who might think that would be helpful. What I don't think would be helpful is having anyone except the doctor or another trained medical professional summarize that information. I do understand that doctors are human and mess up and miss things and might even take worse notes than AI would, but at least it is a Doctor who is doing that. It is a human being who met the other human being and sat in a room with them who is making these decisions.

[–] DarrinBrunner@lemmy.world 3 points 21 hours ago

I expect that doctors agree. Unfortunately, in the U.S. at least, doctors are not in charge of these decisions. Money is. And, money is convinced this will get them more money, so AI is being forced down their throats. Just like everywhere else.

Money is greedy, money is fast, but money isn't smart. We just remember the shit that worked, and forget all the failures along the way.

[–] bluGill@fedia.io -2 points 22 hours ago (3 children)

Codes need to be complicated. Insurance is taking advantage of that (and adding more complexity), but the complexity is in the system in ways that cannot be removed. The common cholesterol test is a lot cheaper to run than a vitamin D test (20 years ago I found online the price list from a lab, IIRC costs ranged from $0.80 to $15,000 - I can't find anything current and my memory might be off a bit, but close enough for discussion), so there needs to be a different code for each test just to ensure the right bill is made. Insurance just uses those codes to decide which they will pay for.

I do not want my insurance to pay for homeopathy or other scams. That just raises my rates and ensures a worse outcome for everyone. So the system of different codes for everything is overall good. There are a lot of debate on how much money we should spend on someone who will die "soon", some would call giving them a quick poison the most humane thing to do - there is plenty of room to disagree on what should or should not be allowed and insurance is just taking a position that not everyone agrees with.

The real problem in my opinion is nobody has a choice about what insurance they have, and so we are all yelling we want something "better" without needing to care about the trade offs.

[–] Akh@lemmy.world 8 points 22 hours ago (1 children)

Insurance… that is the issue.

[–] bluGill@fedia.io 2 points 22 hours ago

Insurance itself is a great idea. However the implementation is all wrong.

[–] dylanmorgan@slrpnk.net 5 points 21 hours ago (1 children)

The government could mandate that all insurance companies use the same codes Medicare does, standardized coding would cut out a lot of actual waste in the healthcare system.

[–] bluGill@fedia.io 2 points 20 hours ago

Again, implementation is wrong, not the idea

[–] MountingSuspicion@reddthat.com 1 points 17 hours ago

I understand different things need different codes, otherwise the labs or other doctors don't know what they need to do, but the fact a doctor can order a test and then a random person with no medical background, or even just a machine, can tell the doctor "no, that's not needed" is a waste in the system and a waste of people's, often sick people's, time. There is no code for homeopathy tests or procedures, so there's no way for them to be prescribed. A doctor can't order 10 ccs of water danced on by fairies, so I don't see that being an issue. My point is that whatever a doctor orders shouldn't need that much oversight. If at the end of the year the government wants to run a "how many MRIs per patient" check on providers they can still do so, but that doesn't require every MRI to be justified in the moment. Then you can investigate the practice for potential fraud if they're not actually doing the MRIs and just charging for them, or if they're being ordered unnecessarily. Doctors should be trusted to do right by their patients. If the government had a going rate for procedures and there was one place where people can see everything that was billed as part of their care things might not need as much oversight.

I admittedly don't work in the field, and I'm aware Medicare/medicaid fraud does happen, but considering the waste (on all sides including patients) created by all the overhead I think we'd come out ahead by just trusting doctors and checking end of year stats.

Also, through the ACA people do technically get quite a lot of choice in their healthcare benefits. I don't think it's a good system because in my opinion your income shouldn't dictate your quality of care or coverage, but if you're looking for more choice in the US and have not looked at the marketplace I recommend doing so. I personally think single payer is better, but definitely look there if you're looking for HSA or deductible differences.

[–] altphoto@lemmy.today 14 points 18 hours ago

MS never had social permission. They just jumped in on the racist asshole government agenda to destroy science and technology entities that worked to protect us from climate change and asshole companies like MS.

[–] november@piefed.blahaj.zone 29 points 21 hours ago (1 children)

I mean, sure, that's a great use case for AI. How about you actually build a machine that can do that instead of pretending LLMs can do it?

[–] undeffeined@lemmy.ml 5 points 18 hours ago

But then they can't take that machine and claim it's the precursor to actual Artificial Inteligence

[–] undeffeined@lemmy.ml 11 points 18 hours ago

These people really live in a different reality than the rest of us.

[–] the_riviera_kid@lemmy.world 17 points 21 hours ago

They never had "social permission".

[–] duckCityComplex@lemmy.world 6 points 17 hours ago

I'm just spitballing here, but maybe you should find out what people want first, and then build that.

[–] Treczoks@lemmy.world 7 points 18 hours ago

The only useful that comes to mind is abolishing it.

[–] NorthoftheBorder@lemmy.ca 8 points 19 hours ago

They never had mine.

[–] gressen@lemmy.zip 18 points 23 hours ago (1 children)

The social permission is an implicit thing at best. It's more a lack of public knowledge about the cost to society.

[–] _chris@lemmy.world 3 points 22 hours ago

Pitchforks and torches?

[–] MushuChupacabra@piefed.world 14 points 22 hours ago

Social permission?

You've never had social permission.

You just force-feed this slop to humanity.

[–] TomMasz@lemmy.world 12 points 22 hours ago

Don't mistake people not being aware of the cost as consent.

[–] mrmaplebar@fedia.io 7 points 20 hours ago

Starting to get nervous that people aren't gonna bite, huh?

That they're still trying to figure out what to do is just further proof that AI is a solution in search of a problem.

[–] ceenote@lemmy.world 14 points 23 hours ago

We need to make the peons irrelevant faster, guys. They're starting to suspect things.

[–] Zephorah@discuss.online 8 points 21 hours ago (1 children)

Wow. No direction just a general plea to “do something useful”.

I still think they’re all competing for the sake of competing. They have all the money and all the credit, what do they care?

[–] bytesonbike@discuss.online 2 points 19 hours ago* (last edited 18 hours ago)

Microslop: "Why don't you do something useful with AI, you idiots?! And when you can, use CoPilot 365 Series XL Pro 11 edition, powered by AI!"

[–] SalamenceFury@piefed.social 10 points 22 hours ago

Yeah, no, you've still got a glorified autocorrect there and blowing more electricity on it won't fix it.

[–] JcbAzPx@lemmy.world 5 points 20 hours ago

So, that is impossible. Just stop already.

[–] sudo@programming.dev 7 points 22 hours ago

Social Permission is quite a new euphemism for shareholder investments.

[–] 20cello@lemmy.world 8 points 23 hours ago

Do something useful, kill yourself

[–] SippyCup@lemmy.ml 6 points 22 hours ago* (last edited 22 hours ago) (1 children)

When a doctor can … spend more time with the patient, because the AI is doing the transcription and entering the records in the EMR system, entering the right billing code so that the healthcare industry is better served across the payer, the provider, and the patient, ultimately—that's an outcome that I think all of us can benefit from.

This is a system that's already in use, doctors aren't doing that work. There's an entire industry around doing that work. If they're going to eliminate that then fucking full send. Single payer or fuck off asshole.

[–] wonderingwanderer@sopuli.xyz 3 points 20 hours ago

It's especially hilarious because he was speaking in Europe...

[–] jessicablaze@lemmynsfw.com 5 points 22 hours ago

Summary: why aren't people throwing money at AI SLOP?

[–] very_well_lost@lemmy.world 3 points 20 hours ago

Since fucking when has the corporate class needed (or even cared about) "permission" to do anything?

Y'all have been raping the environment, eroding our privacy, flaunting regulations and fucking with our elections for basically my entire lifetime.

[–] Mwa@thelemmy.club 2 points 22 hours ago

Just give up tbh