Really lovely article. In paramedicine we usually treat 10g of acetaminophen in a 24-hour window as a potentially fatal overdose. That's also why the law in Australia was changed to require acetaminophen to come in blister packs (harder to get each pill out) of no more than 16. At 500 mg, that only gets you up to 8 g if you eat the whole thing, which is still hopefully non-fatal.
I always thought a simple over-the-counter supplement (NAC) being the cure for an overdose was so cool. It's a pretty cool substance in a lot of ways, and this is a great spur to myself to research it more thoroughly.
I randomly bought NAC just to try it. I dont know about the chemical interactions, but going out with collegues at that time taught me that it's basically impossible to get drunk. Usually a pint of beer is enough to make le feel at least a little dizzy, but when taking NAC, it was all like drinking water
If you all think NAC is great, wait till you try liposomal glutathione (glutathione is one of the things NAC is a precursor for, one of the general take-out-the-trash compounds for your cells). Of all the supplements I’ve tried, it has probably the most immediately noticeable positive effect (maybe because you take it by leaving it under your tongue to be absorbed sublingually for a bit before swallowing). Generally leaves me feeling great, even if I was kind of dragging and tired beforehand.
When I go out drinking with my pharmacist buddy, we take NAC before going out. He swears it makes hangovers less likely. I can't say I've noticed that particular effect, but I do seem to sleep a bit better on those nights.
> Apparently for some people it also helps with lessening tolerance for their ADHD meds, but I'm not so sure about that.
I'd believe it. I first heard of NAC on the nootropic subreddit in a past lifetime. The benefits vary, but generally it's a safe thing with a low chance of making anything worse, but a possibility to improve things. Many neurodivergent folk have written about how they benefit.
I'd give more info on the exact benefits they found (iirc OCD and rumination loops could be broken more easily), but unfortunately my memory is failing me.
My anecdotal experience is that NAC makes me much more tolerant to alcohol. As in, I can drink a lot more without feeling the effects. Since I don't get the same buzz, I care less about reaching for a beer.
How is nac (acetylcysteine) delivered there? I can buy dissolvable tablets here in Europe but from what I see that’s less helpful for mucous, things like mucomyst require inhalation, which isn’t in otc products I know of.
In the Philippines it's available as an effervescent tablet to be dissolved in water. They still tend to work better than the western remedies (guaifenesin etc) even in this form IME.
Usually here in Canada it's available in capsule form which I find less effective.
Same here actually, I find it slightly helpful but the effect’s useful time is limited. I’ve wondered if I could capture the gas released while bubbling and inhale that…
The dissolvable tablets completely fix a runny nose for me. Much better than any nose spray, which tend to irritate the nose and lead to chronic runny nose if taken for too long.
Certain esters have been found to be much safer (in mice, at least):
> The glutathione hepatic values in mice obtained by intraperitoneal injection of the ester are superimposable on controls and the oral LD50 was found to be greater than 2000 mg kg^-1 and the intraperitoneal LD50 was 1900 mg kg^-1 ...
and more general analogs apparently can also be designed to not produce NAPQI:
> Thus, in 2020, N-sulpharyl-APAP prodrugs 39–40
(Fig. 11) were developed. [...] They are not hepatotoxic because they do not generate toxic metabolite NAPQI, even in concentrations equal to a toxic dose of APAP (600 mg kg^−1 in mice).
I am blessed with living in one of the most polluted areas in the world (PM2.5 going into thousands of µg/m³ in winter; summers are not much better due to dense chemical smog). Can you say more about how you're using it to combat that? Thanks!
What does ingesting 10g of acetaminophen even look like? I've got to imagine the fatal dose is far, far, far lower with chronic usage. Finding out that people are ingesting grams is profoundly disturbing.
Extra strength tylenol is 500mg a tablet, so 20 pills. Think most of the accidental ODs are due to people not realizing its included in other cold meds, and also being loopy from whatever illness they're trying to manage, and so end up forgetting when they last took the meds.
Its also a pretty popular choice for people trying to kill themselves, though, so I suspect a non-trival chunk of ODs in the statistics given in the article were intentional.
I've been prescribed slightly more than 5g per day (2 x 650mg tablets every 6 hours) for pain after an operation jointly with ibuprofen, which is scarily close to the limits.
I have taken 4-5g in a day while suffering from intense pain before.
There is a limit to the amount of opioids they will prescribe you, even if you are in mind shattering pain. For instance while attempting to get your dental insurance to actually cover a treatment you may find yourself between risking organ damage or risking $5000+ in ER visit bills only to have them refuse to give you anything but Tramadol.
> I guess it is much better than the situation before that, where you paid $5000+ and they also gave you an opioid addiction.
Having a condition that actually warrants strong opioids and not being able to get them at any price is definitely not an improvement.
The problem is fundamentally that we want to pretend doctors can always distinguish two people describing the same symptoms when one person actually has them and the other is trying to get drugs. The often can't, so you can either make it hard for people to get pain medications even if they need them, or you can make it easy for people to get them even if they don't. And between these the second one is unambiguously better, because the first one is the government screwing innocent people and the second one is guilty people screwing themselves.
> And between these the second one is unambiguously better, because the first one is the government screwing innocent people and the second one is guilty people screwing themselves.
Could not agree more. Depriving people with legitimate pain of opioids is IMHO legitimate torture. It's a bit of a variance on the trolley problem in that the doctor/government isn't causing the pain, but their inaction is prolonging it.
Brother (or sister), you were simply not trying hard enough. I live in a very clean, safe, expensively-policed county, and even I know where to buy fentanyl for much lower cost than a hospital. I would happily turn to that than take 20(!!!) advils in s single day.
Let's review the policy options in light of your suggestion:
1) We make it hard to lawfully acquire pain medications. You pay $$$ to see a doctor and you pay it even if they refuse you. If they do, you then have to pay $$ to get them from Stringer Bell, or start there to begin with if you didn't have $$$, and hope they're not cut with drain cleaner or unevenly mixed so that some days you get 100% corn starch and other days you get a fentanyl overdose.
2) We make it easy. Anyone can get them from Walmart. The people who need them pay the same $ they do for a bottle of Advil/Tylenol instead of paying $$ to murderers or $$$ to waste scarce medical resources that could have saved someone else's life. The bottle from Walmart always has a consistent amount of the drug in it and neither the dental patients nor the addicts get a surprise fentanyl overdose.
Whenever people here mention to my critique of US healthcare how its now mostly solved problem now, its 'good' to see the other side and reality. It certainly doesn't seem solved unless you have a million or two just laying around on the account, while mortgages and kids tuitions are paid. And I can easily imagine a long term condition or 10 which, if unlucky in terms of treatment cost coverage can wipe out that sum in a decade or two, for a single person.
Seriously, how can you guys consider this acceptable. I am not of faith but doesn't bible teach to be kind to your fellow men above all? One would expect more adherence to such basic moral rules in such conservative christian society.
I don’t know a single person in my life who thinks US healthcare is good, so that’s weird. And many my peers a have good jobs with good health insurance. Everyone I know has at least one bad story about insurance, if you’ve ever had more than really basic checkups.
The problem with the US system is that it doesn't know what it's trying to be.
If you did a socialist system then everything is "free" but possibly slow and expensive on the back end when the government isn't efficient.
If you did a libertarian system then everything is cheap but it's caveat emptor because nobody is stopping you from buying morphine for $10 from Amazon.
The US system isn't either one. It pretends to be a market sometimes but then has a bunch of rules to thwart competition. Doctors are required by law to do residency but the government limits the number of residency slots in response to lobbying from the AMA so there aren't enough doctors. "Certificate of need" laws explicitly prohibit new competitors for various services. Insurance is tied to employment to make it hard for individuals to shop around. Laws encourage, require or have the government provide "prescription drug coverage" to make patients price insensitive so drug companies can charge a huge premium for patenting a minor improvement or simple combination of existing drugs and have the patient will something which is marginally if at all better even if it's dramatically more expensive because they don't see the cost when the insurance/government is required to pay for it.
It's a big pile of corruption, because all that money is going to places. But then if you try to fix it, half the population insists on doing the first one and the other half is only willing to do the second one, and the industry capitalizes on this to prevent either one.
Maybe instead we should do both rather than neither. Have the government provide a threshold level of services, like emergency rooms and free clinics and anything more than that the local government wants to fund, and then have a minimally regulated private system that anyone can use if the government system doesn't satisfy them.
I think you're trying to apply ideology where it doesn't belong. Nobody on earth would advocate for such extensive spending to facilitate agreement on financing. It's extremely, extremely inefficient. (But it does produce jobs, which makes politicians super horny.)
The market also won't assist us, as we can't exactly compete future treatment costs against unknown illnesses.
Merely providing emergency rooms and "free clinics" will ensure that people only use these services.
A public option eliminating profit margin seems to at least be sane, and ideally would starve private funding from existence. Any remaining options would highlight deficiencies in the existing system.
A schumpeterian system, if you must slap an ideology on it.
> Merely providing emergency rooms and "free clinics" will ensure that people only use these services.
Emergency rooms operate by triage. If you're having a heart attack, you're going in right now. If your shoulder has been bothering you for six months, you might have to come back multiple days in a row and spend the whole day waiting before there is a slow enough day that you can be seen. There is then an obvious incentive to go pay a private physician to be seen immediately instead. Free clinics are similar: There are no appointments, it's first come first served, and then most people prefer to pay $100 to schedule an appointment rather than wasting an entire day waiting in a queue, but you still have that option for people with no money.
Emergency rooms are also a natural monopoly because in an actual emergency the primary consideration is which one is closest, which doesn't make for a competitive market. So it makes sense to have the government do that. Whereas non-emergency care (which is the large majority of medical expenses) would allow people to compare prices or make cost trade offs against distance or convenience etc., if we would actually expose people to pricing. For example by requiring price transparency and then having insurance pay the second-lowest price for that service within 100 miles of your location, but then letting you choose where you actually want to go and make up any difference yourself, or choose the lowest cost option instead of the second lowest and then put the difference in your HSA.
> A public option eliminating profit margin seems to at least be sane, and ideally would starve private funding from existence.
It's not clear how a government option that doesn't have taxpayer subsidies would do this any better than a private non-profit. There are many existing non-profit healthcare providers and they don't have meaningfully lower costs than for-profit ones.
The general problem is that "non-profits" and government-operated services still have money flowing through them and "profit" can be extracted in all manner of ways other than paying dividends to shareholders. The officers can just pay themselves high salaries, or whoever is in charge of the budget can take bribes/kickbacks to shovel money in the direction of the contractors or unions paying them off.
Meanwhile the nature of "profit" in a competitive market is largely misunderstood because of accounting differences. If a non-profit wants to buy an MRI machine, they have to take out a loan, and then pay back the loan with interest which they account for as an expense. A for-profit company might get the money to buy it by selling shares to investors, and then paying dividends to the shareholders instead of paying interest on a loan, which goes on the books as "profit" instead of interest expense. But you couldn't just replace them with a non-profit and then lower prices by the amount of "profit" they were making because then they also wouldn't have had private investment and you're back to needing the loan and paying the same money as interest to the bank.
The thing that requires providers to be efficient is competition, because then the ones wasting money or taking bribes have to cover the amount wasted/embezzled by charging more to customers and then the customers don't choose them because they have higher prices. But that's the thing the existing regulatory system goes out of its way to thwart.
Buy a pack of 20x500mg (just checked, common size in Germany), take 2-3 every half hour for a while.
Sure, that's extreme. But if you're unaware of the risks, you feel sick, and you believe it's helping you.
I mean, people aren't killing themselves in masses with it, but it happens every now and then. Easily imaginable that one in a few million people will have enough tendency to take more pills and is unaware of the overdose danger.
Taking too much acetaminophen is bad for you but 10g is 20 extra strength pills and that much isn't likely at all to kill you but damage your organs is quite possible. Reading this might make someone in a bad place think that much will do the job and it won't. Tylenol poisoning's most likely outcome is permanent organ damage and pain, don't try it.
As an American this is such a weird question to me. I purchase my ibuprofen and benadryl in bottles of anywhere from 400 to 1000 pills every few years.
Apparently a common source of problems is taking two different medications without realizing they both contain acetaminophen.
Suppose your arthritis is acting up, so you start taking Tylenol 8hr Arthritis Pain[1]. That's 2 tablets every 8 hours. They're extended-release with 650mg per tablet. A total of 3900 mg in 24 hours.
A few days later you get the flu, so you decide to add what seems like a completely different medication: Theraflu Flu Relief Max Strength[2]. It has a cough suppressant and an antihistamine. But each caplet also contains 500 mg of acetaminophen. It says to take 2 caplets every 6 hours, so you take 8 of them in 24 hours[3]. That's another 4000 mg.
Between the two, you're at 7900 mg.
Then you wake up in the morning and take both medications, but 30 minutes later you've forgotten you took them. You're not thinking straight because you're sick. So you accidentally take a second dose. That additional 2300 mg brings your total to 10200 mg.
[3] You weren't supposed to take 8 of them, though. If you'd read the label very carefully, you'd have seen it also says not to exceed 6 in a 24-hour period.
My personal rule is to only purchase over-the-counter meds with a single active ingredient. I'd rather separately take an antihistamine, expectorant and painkiller than a concoction where I have to read the whole label and do math while sick to separate the doses and timings.
There are some that are very hard to find as a single ingredient. Recently I was purchasing a medication for back pain, I had a choice as to which other ingredient I wanted, but I didn't have the choice of none. I picked the combined ingredient I don't like to take, because I wouldn't be adding it on top.
I did toss on the other option, stand alone, at one point so I could get some sleep.
It left the medication I was more comfortable taking as an add-on option if things got bad enough. (This particular medication has much lower risk of overdose, so if I got stupid and took it again there would be no significant additional risk.)
It's ironic, but taking the combined medication with a known higher risk of its own was better than taking the lower risk medication.
One was controlled, higher risk, taken at specific times, while the other was taken in addition, on demand, as required.
Specifically this is one reason they’ll sell you cocodemol or Vicodin but not codeine or hydrocodone directly — if you take enough to get a codeine high, you’ll have taken a toxic amount of paracetamol/acetaminophen, so they assume you won’t.
I didn't until I had a bulging lower back disc pressing on my sciatic nerve. My leg felt like it was constantly on fire no matter what position I put myself in. In the past I've torn my ACL and had surgery to reconstruct and that pain was like stubbing my toe compared to the back pain. I understood how people become addicted to pain meds after my back situation.
Totally get it, I too only understood it "theoretically" till I had a (fairly minor!) dental operation.
... Suddenly I'm maintaining a continuous note of when I'm taking which medicine to avoid crossing safe limits (which I anyway was crossing most days).
I was only told to take 2 paracetamols a day (bullshit dose, I'd be waking up from the pain even with more pain meds).
"Diclofenac for rare use" - well, if nothing else is touching the pain, is it an emergency?
Eventually after forever I was able to transition to Ibuprofen + paracetamol. And I already have a health condition which is heavy on my kidneys... pain management can be absolutely crazy.
Pain management can be crazy but in your case it sounds like they simply didn't prescribe the appropriate medication presumably due to the anti opiate hysteria that has taken hold.
While that's quite possibly true, I forgot to mention that I'm not in the US but India. I was conscious the whole time, with only local anesthesia. Also the dentist in question is actually our "family" dentist, and he's a pretty knowledgeable/skillful guy (easily more knowledgeable than many GPs on health matters of the body).
Fun fact, you can totally get them to pause the procedure without saying a word. All you have to do is end up in a lot of pain, have your heart rate skyrocket like anything, and get everyone in the OT very concerned ;)
I had severe nerve pain due to a herniated disc. While awaiting a surgery, I was prescribed an opioid (Tramadol) but it didn't seem to help much at all. Acetaminophen actually worked better than the opioid for me...
This can easily happen over the course of 24 hours if you're in "fuck me I'll do anything to make it stop" levels of pain. I've taken more than 20 ibuprofens in a day a few times in my life, which, while not medically advised, did not kill me. I actually had no idea acetaminophen was so dangerous.
Just in case, ibuprophen does not work well for pain relief [at lest for some kind of pain]. Paracetamol [acetaminophen] usually is much better against pain.
And paracetamol + ibuprophen can help with strong pain for which neither paracetamol or ibuprophen work at normal doses.
Not really. Both address different sources of pain, and do so using different processes.
Ibuprofen is a Nonsteroidal Anti-inflammatory Drug (NSAID) that reduces pain and inflammation, while acetaminophen does not. (Acetaminophen is believed to act mainly in the brain rather than at the site of injury).
Ibuprofen- Fundamentally, if the pain is caused by inflammation, reducing the immune systems response to it can reduce pain, but if the pain is more acute it won't make a dent.
With acetaminophen, taking more isn't a solution in most cases, you need another method to reduce the pain further if it doesn't achieve its goal.
(That's why it's combined with things like codeine, which affects the brain in a different way for an additive effect)
> you need another method to reduce the pain further
I don’t know about “most cases” but often you don’t want to reduce the pain _further_, you want to reduce the pain _again_. (Having an alternative definitely helps in the meantime.)
The .nl indicates the netherlands. Many people in the netherlands vent/joke about how the doctors here only ever tell you to take paracetamol and come back in two weeks if it's still a problem (recursive solution).
However the last time I went to my GP she scoffed at me taking the maximum and suggested I take literally double the maximum recommended dose 4-5 times a day which totaled I think 2.5x the daily maximum on the package. I am very much a "believer" in science and reasonable medical authority but this experience sowed the seeds of doubt, because from what I have always heard, that can actually kill you or cause permanent liver issues. I was also taking diclofenac simultaenously, and when I told her how many mg, she asked "where can you even buy such small doses, that's what I would give a small child" =/
My understanding is double the max dose of paracetamol is the LD50. Seems crazy for your GP to advocate for that. Published recommended dose is assuming average weight, so maybe if you're a very large person, the advice makes some sense, but wow.
They are common in France, but not in such packages: There are restrictions that prevent you from buying more than than 8g/day (theoretically at least, I don't believe they are strictly applied in practice).
They are widely sold at 1000 mg (1g) tablets in Europe, but in many countries they require doctor's prescription. There are also purchase limits to the number of pills you can buy at once.
I've heard it suggested that acetaminophen just come with a small dose of NAC alongside it to make it safer. I guess this would require a lot of regulatory work to approve, but given that 500 people a year OD, it seems like a thing we should at least consider.
Meanwhile, it's funny that it seems like acetaminophen should safer in more scenarios, but the other has a lot of overdoses with typical use, I guess that's why there's a gap between the two, because ODs are apparently a lot more common or at least more legible than problems caused by the other drug.
You can still buy 100 packs, they are just behind the counter at chemists. TBH it's a rather stupid restriction - do they think people only ever own 1 packet of paracetamol at a time? In my household we have at least half a dozen, including a 100-pack from Oz and a 500-pack from America.
Oh right - that's probably what we did, buy a big pack from behind the counter.
I don't think you can even do that in the UK.
Yeah we usually have a few packs hanging around, and I get the 'it seems stupid' thing, but sometimes just adding a tiny bit of friction when someone's trying to kill themselves might save a life. I dunno, I hope that's shown in the evidence anyway. Otherwise it's just pointless like the whole pseudoephedrine song and dance, which has inconvenienced anyone looking for a decongestant while doing sweet FA to the availability of meth.
> Oh right - that's probably what we did, buy a big pack from behind the counter.
No, when you visited they were still on the shelf. They only put them behind the counter in 2025.
> sometimes just adding a tiny bit of friction when someone's trying to kill themselves might save a life
I'm philosophically not for making suicide harder. If someone wants to die, that's their right. And practically, while you might be able to show a stat-sig decrease in paracetamol poisoning, I'd expect the suicides to largely just move to other methods.
The point is that many don't really want to. Those that actually want to can buy two boxes from two shops or ask the pharmacist for the big pack from behind the counter.
This just adds a tiny amount of friction to impulsive attempts, which may be a classic cry for help or just someone in the depths of some sort of mental health episode. Such folks may think better of it the next day and a very small amount of inconvenience will put them off. I think suicide is serious enough that you should probably mean it, and societally saying 'think twice about this' is a good thing.
On the idea that it just shift deaths, as your sibling poster points out (from the UK) -
"in the 11 years following the legislation there were an estimated 765 fewer suicide and open verdict deaths from paracetamol poisoning, which represented a reduction of 43% [...] This reduction was largely unaltered after controlling for a downward trend in deaths involving other methods of poisoning and also suicides by all methods."
So it looks like this tiny, tiny barrier does actually deter people. And that definitely points to them not really being sold on it in any rational way.
I just don't buy the paternalism. People have free will, if they want to do something they would regret later, it's still their right.
That quote doesn't say what you think it means. It's not talking at all about whether suicides shifted to other methods; it only says that there was a secular decline in poisonings (-32%) and suicides in general (-10%) during the study, so they have to also discount some of the raw 48% drop in paracetamol as being part of that broader trend and not due to the treatment. They come to the 43% number only with a generous assumption that had the law not gone into effect, there would have been an increasing trend in deaths from paracetamol poisoning, which seems wrong to me. The more obvious way to derive the prior would be to look at non-paracetamol poisonings and expect the same trend, in which case the effect might be something like -24%.
Anyhow, it's still perfectly possible that the people who were deterred from paracetamol poisoning committed suicide some other way; the data in that paper says nothing about it.
> People have free will, if they want to do something they would regret later, it's still their right.
Then this minor frictional measure is the very least of your worries. For a start, any given pharmacy has an entire pharmacopoea of compounds that people are kept away from for their own good. Not to mention liquor licensing rules making landlords cut folks off at a bar if visibly drunk etc. And guard rails to stop people climbing to high places. And ... preventing people from doing stupid shit in the moment is everywhere in our societies.
There are a heck of a lot of things I'd put higher up my list of concerns than "may have to visit two shops if wanting to kill myself"
Paraphrasing from [0], after September 1998 when the restriction was introduced, "The annual number of deaths from paracetamol poisoning decreased by 21% [...] the number from salicylates decreased by 48% [...] Liver transplant rates after paracetamol poisoning decreased by 66% [...] The rate of non-fatal self poisoning with paracetamol in any form decreased by 11%"
See also [1]: "in the 11 years following the legislation there were an estimated 765 fewer suicide and open verdict deaths from paracetamol poisoning, which represented a reduction of 43% [...] This reduction was largely unaltered after controlling for a downward trend in deaths involving other methods of poisoning and also suicides by all methods."
Yes, and you can still die in a car crash if you're wearing your seatbelt, and wearing a helmet on your motorcycle won't save you from a head-on with a truck, and you can still drown in a pool with a lifeguard, and you can still die in a burning building with smoke detectors.
Harm reduction is about shifting probability distributions, not guaranteeing outcomes. Kids can still get into pill bottles with childproof medication caps, but accidental ingestion of aspirin by children reduced by 40-55% after they were mandated. [0]
No. Ethanol and tylenol compete for CYP2E1 that produces toxic NAPQI, so no, acute alcohol intoxication has a protective effect at least where it comes to tylenol toxicity.
Alcohol and Acetominophen/paracetamol should not be mixed.
When alcohol enters the picture, it increases the activity of CYP2E1, so the body produces more of the NAPQI toxin. Alcohol also decreases glutathione production, the body’s natural defense mechanism, meaning NAPQI is more likely to build up in the liver in dangerous concentrations.
There is a danger in chronic abuse resulting in upregulation. Mixing the two at once is no problem for the liver, which is also why patient information leaflets for paracetamol do not contain a warning to avoid alcohol, only about chronic alcohol abuse.
Your crappy source is vague in what consumption pattern constitutes a risk and actually cites a better source that supports the idea that acute alcohol consumption reduces paracetamol toxicity. https://www.biorxiv.org/content/10.1101/2020.07.07.191916v1....
That's a mathematical model, but this relationship between the two is what I was taught in medical school and it is still supported by the science. There's plenty of other sources, I just picked that one because your article cites it. Just search for "paracetamol ethanol" on Google Scholar.
Growing up we learned about _Slaughterhouse 5_ and _Cat's Cradle_ by Kurt Vonnegut. But there's not enough discussion or awareness of _Player Piano_. Incredibly prescient. These kinds of dystopic headlines are exactly the kind of thing you'd see in the book.
Player Piano is a 1952 Sci-fi novel by Vonnegut which explores the social and economic impact of automation replacing labor. If I recall correctly (I read this 15+ years ago) it is told from the perspective of one of the last people with an actually useful job, a person who's job it is to fix the machines that automated away jobs.
Paper Computing (great name!) is something I've been thinking about a lot to help my kids benefit from tech without exposing them to the brain melting addiction of screens. I sacrificed a few crazy nights of sleep to try to build a Paper Computer Agent prototype for a recent Gemini hackathon (only to disappointingly have submission issues right before the actual deadline) which my kids loved and keep asking me to set up permanently for them.
It's essentially a poor man's hacked up DynamicLand - projector, camera, live agent. There are so many things you could do if you had a strong working baseline for this. My kids used it to create stories, learn how to draw various things, and watching safe videos they could hold in their hand.
There's something weirdly compelling and delightfully physical about holding a piece of paper that shows a live rocket launch, with the flames streaming down the page. It could also project targeted pieces of text, such as inline homework advice, or graphs next to data. It doesn't take long to imagine any other number of fun use cases, and it feels a lot more freeing and inspiring than keeping everything bound to a screen.
R.I.P. to the Amazon Glow video calling device, killed before AI went mainstream. I'd love to hear how to get root on one... exactly the hardware your project could use most effectively and an amazing interface for playing games remotely with the grandparents.
I love how creatively ai is integrated in here. Amazing.
The Folk Computer people have some incredible work they've been doing too, that's definitely worth looking at for anyone interested. Their intergation of a novel display technology is really sweet too, allowing for good visibility in a variety of conditions, which I love. https://folkcomputer.substack.com/https://folk.computer/https://news.ycombinator.com/item?id=39241472 (165 points, 2 years ago, 53 comments)
I love gow creatively ai is integrated in here. Amazing.
The Folk Computer people have some incredible work they've been doing too, that's definitely worth looking at for anyone interested. Their intergation of a novel display technology is really sweet too, allowing for good visibility in a variety of conditions, which I love. https://folkcomputer.substack.com/https://folk.computer/
Thanks for all the kind words. I would love to work on this more but the hackathon sprint was really all I had time for (note the newborn in the video...) without more backing / support. I was really bummed that the hackathon rejected the submission, because it provided some Google support if you win.
If anyone knows of a way to develop this... the code is on Github, and I have a roadmap in mind, but as we all know there's a huge gap between hacky prototype and "works smoothly for other users".
this is really cool, I'd love to use something like this for my kids too. Maybe I'll try your project when I have some more free time. Would love to contribute but i'm not very skilled in python.
If you don't mind me asking, what hardware did you use? Especially for the project, I'm guessing it needs to have quite a strong bulb in order to be seen in broad daylight?
I was pretty excited when I saw the premise behind what Apple was doing with VisionPro because I figured they were steering towards this, but it seems they’ve looked away and don’t really care about going deeper into this direction.
I asked at some point if I could theoretically develop an application that could literally be controlled by a Fischer Price toy, like a little plastic car console or something. Or even potentially have a real keyboard that isn’t connected to anything, but the VisionPro can just see my keypresses and apply them as if I was actually pressing something. The former case is possible, but surprisingly difficult, but the latter case isn’t really there yet (requires too much precision and latency is worse than just using a Bluetooth keyboard).
Either way, the idea of a computing environment that meshes with and directly interacts with the real, physical objects around you is an interesting premise I’d like to see taken further with “Spatial Computing”/AR. Scanning and recording things I’m writing on a whiteboard or in a notebook by recognizing that I’ve picked up a pen and am writing something down would just be getting started.
Of course, if we’re ambiently recording everything you’re doing there will need to be some kind of regular process/interface to “sift” everything at the end of the day. This is the core of the Getting Things Done methodology. Everything goes into a big “intake list” and then you do periodic check-ins throughout the day where you review the list and decide whether to move those to a series of sub-lists to “do this now,” “do this soon,” or “do this someday.”
If you keep this up, we're going to have to ban you. I don't want to ban you, so if you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.
I think better phrasing such as "Have you considered the ramifications of exposing and normalizing AI for your children?" That is a lot more pleasant than verbally bashing someone with a club like a cave man (Though I get it. sometimes, "og want smash." but keep og in check.)
Thanks, this might be exactly what I'm looking for.
I see you have support for vanilla js and svelte, but it's unclear whether you can get all the same functionality if you don't use React. Is React the only first class citizen in this stack?
> Is React the only first class citizen in this stack?
Each system gets the same functionality. We centralize the critical logic for the client SDK in "@instantdb/core". React, Svelte, Tanstack, React Native et al are wrappers around that core library.
The one place where it's lacking a bit is the docs. We have specific docs for each library, but a lot of other examples assume React.
We are improving this as we speak. For now, the assumption on React is quite light in the docs, so it's relatively straightforward to figure out what needs to happen for the library of your choice.
Any thoughts on a potential Tanstack DB integration?
I love tanstack's rich front end. Ya'll have quite an amazing system, and I'm wondering if there's any thoughts on how perhaps your pretty substantial front end side might be adapted to tanstack DB. https://github.com/TanStack/db/tree/main/packages
Thanks, this helped crystallize something for me: the play the AI labs are making is anti-fragile (in the Nassim Taleb sense):
> The very act of resisting feeds what you resist and makes it less fragile to future resistance.
At least along certain dimensions. I don't think the labs themselves are antifragile. Obviously we all know the labs are training on everything (so write/act the way you want future AIs to perceive you), but I hadn't really focused on how they're absorbing the innovation that they stimulate. There's probably a biological analog...
Well there are many, and I quote this AI response here for its chilling parallels:
> Parasitic castrators and host manipulators do something related. Some parasites redirect a host’s resources away from reproduction and into body maintenance or altered tissue states that benefit the parasite. A classic example is parasites that make hosts effectively become growth/support machines for the parasite. It is not always “stimulate more tissue, then eat it,” but it is “stimulate more usable host productivity, then exploit it.” (ChatGPT 5.4 Thinking. Emphasis mine.)
Instead of anti-fragility, I'd point you to the law of requisite variety instead.
You'll notice that all AI improvements are insanely good for a week or two after launch. Then you'll see people stating that 'models got worse'. What happened in fact is that people adapted to the tool, but the tool didn't adapt anymore. We're using AI as variety resistant and adaptable tools, but we miss the fact that most deployments nowadays do not adapt back to you as fast.
New models literally do get worse after launch, due to optimization. If you charted performance over time, it'd look like a sawtooth, with a regular performance drop during each optimization period.
That's the dirty secret with all of this stuff: "state of the art" models are unprofitable due to high cost of inference before optimization. After optimization they still perform okay, but way below SOTA. It's like a knife that's been sharpened until razor sharp, then dulled shortly after.
> If you charted performance over time, it'd look like a sawtooth
People have, though, and it doesn't show that. I think it's more people getting hit by the placebo effect, the novelty effect, followed by the models by-definition non-determinism leading people to say things like "the model got worse".
Is this insider info? The 'charted performance' caught my eye instantly.
Couple things I find odd tho: why sawtooth? it would likely be square waves, as I'd imagine they roll down the cost-saving version quite fast per cohort. Also, aren't they unprofitable either way? Why would they do it for 'profitability'?
It's rumors based on vibes. There are attempts to track and quantify this with repeated model evaluations multiple times per day, this but no sawtooth pattern has emerged as far as I know.
I don't want to go too far down the conspiracy rabbit hole, but the vendors know everyone's prompts so it would be trivial for them to track the trackers and spoof the results. We already know that they substitute different models as a cost-saving measure, so substituting models to fool the repeated evaluations would be trivial.
We also already know that they actively seek out viral examples of poor performance on certain prompts (e.g. counting Rs in strawberry) and then monkey-patch them out with targeted training. How can we be sure they're not trying to spoof researchers who are tracking model performance? Heck, they might as well just call it "regression testing."
If their whole gig is an "emperor's new clothes" bubble situation, then we can expect them to try to uphold the masquerade as long as possible.
It's not insider info, it's common knowledge in the industry (Google model optimization). I think they are unprofitable either way, but unoptimized models burn runway a lot faster than optimized ones.
The reason it's not a square wave is because new optimization techniques are always in development, so you can't apply everything immediately after training the new model. I also think there's a marketing reason: if the performance of a brand new model declines rapidly after release then people are going to notice much more readily than with a gradual decline. The gradual decline is thus engineered by applying different optimizations gradually.
It also has the side benefit that the future next-gen model may be compared favourably with the current-gen optimized (degraded) model, setting up a rigged benchmark. If no one has access to the original pre-optimized current-gen model, no one can perform the "proper" comparison to be able to gauge the actual performance improvement.
Lastly, I would point out that vendors like OpenAI are already known to substitute previous-gen models if they determine your prompt is "simple." You should also count this as a (rather crude) optimization technique because it's going to degrade performance any time your prompt is falsely flagged as simple (false positive).
I've never yet been "that guy" on HN but... the title seems misleading. The actual title is "A Ramsey-style Problem on Hypergraphs" and a more descriptive title would be "All latest frontier models can solve a frontier math open problem". (It wasn't just GPT 5.4)
It's AI narrated, but at this point if I heard Zvi's actual voice I think I would be confused. It's really well done, and uses different voices for each new person being quoted. It also has really good narrated image descriptions.
Zvi's articles are literally exhaustively long,l - before I was able to listen to them I got tired trying to read the whole thing. Now it's my favorite way to keep up with AI.
I've been thinking about sovereign AI a lot lately. About a year ago I was wondering what each country would be doing, and looking at places like e.g. Australia (which has pretty strict data residency laws for certain industries) - at that point I thought about advocating for why such countries should train their own models, but now I'm having a harder time justifying that point.
I can't see how any of these other countries could even approach the level of capability of the big three providers. I can imagine only a handful of countries who could even theoretically put enough resources towards reaching the SOTA frontier. Sure, even a model of capability level ~2024 has plenty of valid use cases today, but I'm concerned that people will just go with the big three because what they offer is still so so much better.
Not trying to discourage efforts like these, but is there really a good case for working on them? Or perhaps there's a state/national case, but it's harder for me to see a real business case.
India has a lot of languages and people need access to something than allows them to do basic stuff with it. I don't think relying on the US is a long term solution.
An example. I am into proofreading and language learning and am forced to rely on Claude/Gemini to extract text from old books because of the lack of good Indian models. I started with regular Tesseract, but its accuracy outside of the Latin alphabet is not that great. Qwen 3/3.5 is good with the Bombay style of Devanagari but craps the bed with the Calcutta style. And neither are great with languages like Bengali. In contrast, Claude can extract Bengali text from terrible scans and old printing with something like 99+ percent accuracy.
Models specifically targeted at Indian languages and content will perform better within that context, I feel.
Seems like you and the author are doing the same thing: speaking in absolutes. It's possible for "Anthropic" (or the summed vector of all the human decision makers within it) to have contracted with the military because it wants to make money AND it wants to help.
The questions are: "Help with what, precisely?" and "How much money versus how much value (/principles) compromise?"
I've worked for big corporations for a long time, and one of the first things I've learned is that individual motivations mean very little, if anything. At the end of the day, the bottom line is all that matters. And we know this is particularly true of big LLM companies given their track records.
I always thought a simple over-the-counter supplement (NAC) being the cure for an overdose was so cool. It's a pretty cool substance in a lot of ways, and this is a great spur to myself to research it more thoroughly.
reply