There was a degree of proof of work involved. Text took human effort to create, and this roughly constrained the quantity and quality of misinforming text to the number of humans with motive to expend sufficient effort to misinform. Now superficially indistinguishable text can be created by an investment in flops, which are fungible. This means that the constraint on the amount of misinforming text instead scales with whatever money is resourced to the task of generating misinforming text. If misinforming text can generate value for someone that can be translated back into money, the generation of misinforming text can be scaled to saturation and full extraction of that value.
I do not think it true that it goes "both ways." Patients do not have a collective responsibility to doctors. Patients should not go untreated because other patients lie about their addictions. Being a doctor means exactly watching dozens of horses go in and out the door each day and saying "that's a zebra" before the zebra makes it to the exit. There's a fundamental responsibility to perform differential rather than normative diagnosis.
Patients do not have a collective responsibility to doctors, but I dont think doctors have the responsibility to provide a diagnosis at all.
I think people have strange and exaggerated expectations.
You don't go to a car mechanic and expect them to have all the answers and perfect accuracy. The expectation is that they will take a look and provide their fallible opinion.
What standard of care are you proposing? Please be specific.
Auto mechanics frequently misdiagnose problems, especially those caused by electrical or software faults. But in the worst case they can usually just keep following the manufacturer's service manual and replacing parts in a trial-and-error process until the vehicle works again.
The human body is orders of magnitude more complex and there is no service manual. We have a few evidence-based medicine clinical practice guidelines but those cover only the simplest of cases. For anything more complex, physicians have to fall back on theory, intuition, and experience. It's not surprising that they sometimes get it wrong. And sometimes there's just no way to make a definitive diagnosis for the root cause of a patient's complaints and so treatment is necessarily symptomatic; this can be tough for patients to accept. I'm not trying to defend clinicians who make preventable errors or dismiss legitimate patient concerns but we need to be realistic about what is achievable given the current limited state of medical knowledge.
The standard here was the doctor didn’t listen to the woman. He could easily have found she was lying or not with a simple overnight visit to the hospital. Put her in there for one night, see she’s not drinking but still drunk, and that’s it. Instead, she suffered for ages.
Listening is a simple standard. Doctors don’t listen because they don’t care further than getting more patients through the door faster.
If I tell my mechanic there is a problem with the steering, he’s not going to change the oil and send my car out, he’ll check the fucking steering.
Did the woman come in and say "I have auto-brewery syndrome, and I want you to test me for it"?
If so, then the comparison is off. It isnt that the doctors "didn't listen", it is that they didn't correctly deduce a 1 in a million cause, based on the information they had.
She came in, the doctor asked if she had drunk alcohol, she says no, what more is there to understand? This doctor didn’t listen. If he had, even if he didn’t know about this specific disease, he could have started tests and brought in other doctors. Instead, suffering. Not hard to understand. Doctors don’t listen.
You keep saying they don't listen, but there's no evidence they didn't hear what she said. It seems your problem is more with how much weight they put on that information, and how much effort they put into getting to the bottom of things.
Most doctors aren't interested in playing Detective for the extremely rare cause. They treat the most likely cause given the information that they have on hand
I think you are hung up on the word listen, and I am saying there are a lot of things that happen after they hear what a patient says.
patients aren't saying "I have auto brewery and would like you to confirm it".
Patients are instead reporting symptoms which the doctor then has to interpret and find a likely cause. Even if they 100% believe the patients, the diagnosis may not be obvious. IF they dont 100% trust the patient, or think they may be confused, then it is even harder. Patients ARE very unreliable.
They think she’s a drunk, she says she isn’t, the easiest way to find out is ensure she can’t have alcohol. Not that hard to do at a hospital! Seems like a simple idea to me.
Remember, it took several visits to figure this out, and no one even tried this simple thing to suss out if she was truly an alcoholic.
My standard is that doctors listen to patients, which they don’t do because they don’t respect them. They think they’re all knowing, incapable of mistakes, and let their personal biases rule. Every doctor I’ve been to has struggled to listen.
I have never expereinced a doctor that doesnt listen to their patient. However most dont take everything the patient says at face value.
Processing, weighing, and interpreting what patients say is a fundamental part of their job. Sifting through crappy data and figuring out what is relevant.
Severe auto-brewery has been documented on the order of 20 times in the history of the western world. Thats why it is worth publishing a paper about, and why doctors dont suspect it, and might not even know about it.
How many people have have gone to the doctor and lied about their alcohol use? 6 nines is 1 a million patients. There are probably more nines than that.
When there are single digits for the number of people experiencing this, and many digits for those who are alcoholics, you cannot expect anything but Occam’s razor.
Generally speaking, I defer to your expertise point of view in the matter, and I agree that it will be far easier to generate meaningless research that passes the test of appearing meaningful to reviewers than it will be to generate meaningful research that passes the test of appearing meaningful to reviewers.
However, it is an open secret that this is already true, is the thing. Meaningful peer review is already confined to islands within a system that has devolved into generating content. The automation of the process doesn't represent a tipping point, and I don't think that the ethically disclosed production of 'research' by large language models is going to represent a significant part of the problem. The errors of the current system will be reduced to absurdity by the existent ethical norms.
There's an interesting story inside this interesting story. In the counterfactual, the company continues to manufacture and sell the product as a cordless telephone for a few years before the rest of the vendors catch up with it in terms of quality. That's not what happened. Why not?
I think the question on everyone's mind is probably how to produce a signal from inside the hull of a single-hulled cargo vessel in the North Atlantic that will allow one to relay messages in morse code to the US Navy.
There was a shift in how the word 'hominid' was used around the 1990s, largely due to the introduction of genetic classification if I understand correctly. Hominids (humans and close ancestors) used to be considered distinct from Pongoids, which were basically all other great apes including chimpanzees and bonobos. That taxonomy was retired, and all great apes are now considered hominids.
I don't think that it would be wise for anyone to rely on such practices. Even with the best of intentions, obsolescence and unintentional misdirection are strong possibilities. Considering normative intentions, it is an invitation for "optimization" attempts by websites presenting contested information.
But maybe you guys are 1.25 developers. I mean if most people are staring at their screens and done for the week on Friday and you're still getting things done then in terms of ideals — because okay I know this needs hashing out in practice— I don't think anybody is going to resent you taking home a premium based on what you put in every week.
Honestly if knowledge workers are sort of dialing it in unhappily with minimal productivity after X hours and labor/service/on-call workers can keep plugging away for Y>X hours then there might be some win-win there in terms of quality of life and wage inflation.
The charges will not be like nitroglycerin, but they will be highly explosive and will detonate when sufficiently energized. Under the right circumstances the kinetic energy from a fragment of an interceptor would suffice.
Most modern high explosives won't detonate at all in response to a kinetic insult. They need a detonator, and usually a primer too. Even TNT; you can hit it with a hammer, shoot at it, or try to set it on fire; it won't explode. To set it off, you need a supersonic shockwave.
To my circumstantial understanding, it is plausible that fragments of an interceptor detonated in proximity to the warhead could achieve supersonic velocities capable of delivering the requisite shock, but I lack the knowledge to argue the point.
reply