I think 100 years after their death would be reasonable because at that point it's long enough that people won't assume there's an actual connection to the person or that it's endorsed/founded by them
Indeed. Just knowing that the subjects who followed through with the shocks were less likely to obey the rules could be interpreted in many ways, some invalidating the results of the experiment, some just suggesting a mechanistic explanation, and some making the results even more concerning.
* Did the subjects who went full voltage stop caring about the "learning" protocol because they realised it was all fake? Then the conclusions of Milgram's experiment are invalid.
* Did the subjects who went full voltage make more mistakes because they were more anxious and fearful of the experimenter? Then underlying fear might be a mechanism for blind obedience, and further research would be interesting.
* Did the subjects who went full voltage just enjoy electrocuting the dude so much that they stopped caring about asking the questions correctly? Then blind obedience is the least of our worries, widespread sadism is much more concerning.
It says more then that. The "psychopats" were NOT following the rules. The rule followers were not cruel.
The act of torturing was not due to the torturer obeying the rules. Instead, torturers broke the rules and created conditions that allowed them more torture.
It's pointless to write a whole article about how model collapse is actually happening and isn't just a theoretical concern with no evidence that model collapse is actually happening.
The key missing step which breaks the loop is that while indeed a larger and larger portion of the web is written by language models, that data isn't being used to train new models - at the beginning of LLMs people did indeed want to use "all the web" to train models, but that's not being done now anymore, you either take only old pre-LLM data, or you pay for new 'clean' data, or take extensive filtering steps to avoid accidentally ingesting synthetic data.
The main phrase of the title "model collapse is happening" is untrue and not substantiated in the article - all the true statements in the article are about the hypothetical problem, warning of the bad consequences that would likely happen if makers of major models did something they aren't doing, but they aren't doing that because that is a known issue that they're avoiding. It's like writing an article "Foot shooting epidemic is happening" with a long, solid (and true!) proof that if you'll shoot yourself in the foot, it will indeed cause serious injury...
> It's pointless to write a whole article about how model collapse is actually happening and isn't just a theoretical concern with no evidence that model collapse is actually happening.
Except perhaps the link to article on the peer-reviewed paper that describes the problem in detail.
> Researchers at Oxford and Cambridge published work on this back in 2023, showing how iterative training on synthetic data leads to progressive degradation.
He is going around talking about biblical prophecies and the antichrist so regardless of whether you consider thiel a Christian or whether he considers himself a Christian the comment you are responding to is entirely accurate in saying he is interested "only in the hate and doom parts of [Christianity]".
Peter grew up in an Evangelical household which probably shaped his framing/worldview. As an adult he still identifies as a 'heterodox protestant'. Which in America usually means he's not really Christian, he just picks and chooses some ideas from it. The way he uses 'Antichrist' to talk about politics and tech (not just a single person or religious entity) seems to confirm that idea.
I think you are correct that the headline is wrong because Nintendo is not asking to be refunded $200 billion, but to be precise the number doesn't seem to be the compensation sought by all the companies that are suing either.
Nintendo is the only plaintiff in this specific suit so joined in this article does not mean they are all plaintiffs in one suit, just that they have all filed suits.
The Nintendo complaint specifically mentions $200 billion, but I think this is just supposed to be the rough amount the us government collected in total based on the tariffs that the supreme court found to be illegal from all companies/countries (as background information), not what Nintendo or any other specific companies that are suing are requesting be refunded.
> Better air quality is one of the best (and cheapest) ways to improve overall health.[1]
While I think that the things the person lists on that site are good precautions (although I think not using an ultrasonic humidifier would be better stated as only using distilled water with ultrasonic humidifiers), I think that saying that doing these things is "is one of the best (and cheapest) ways to improve overall health" is too strong, because while there are lots of studies showing population level correlations between pm2.5 particles and health problems, I don't think there is currently evidence specifically showing that things like using air purifiers actually improve health.
>I think not using an ultrasonic humidifier would be better stated as only using distilled water with ultrasonic humidifiers
That's a common misconception. They still atomize bacteria from the tank because nobody ever cleans & sterilizes their dehumidifier tank often enough. Distilled water users still report high indoor particle counts when it runs, which is how you know it's an ineffective prevention.
The cost of distilled water every year (even making it at home) means ultrasonic humidifiers are the most expensive option too.
>while there are lots of studies showing population level correlations between pm2.5 particles and health problems, I don't think there is currently evidence specifically showing that things like using air purifiers actually improve health.
Air purifiers reduce PM2.5 concentration, so if PM2.5 is bad for health then air purifiers are good for health.
This is like "there's no evidence parachutes improve survival after jumping from a plane."
> Air purifiers reduce PM2.5 concentration, so if PM2.5 is bad for health then air purifiers are good for health.
> This is like "there's no evidence parachutes improve survival after jumping from a plane."
A proposed intervention like air purifiers that have no direct evidence but are plausible because they reduce something that has been shown to be correlated with negative long term health outcomes is the opposite of something like parachutes that we directly know work.
This is a completely bizarre comparison and it's like saying that skepticism of an Alzheimer's drug targeting amyloid plaques is like doubting parachutes work because we know that amyloid plaques are correlated with Alzheimer's.
> reduce something that has been shown to be correlated with negative long term health outcomes
Are you suggesting the link isn't causal? Because your argument only makes sense if you think air pollution particles aren't actually the cause of lung damage, which is the opposite of the scientific consensus.
The fact that you have to bring up amyloid plaques is especially a red flag, since this is famously a rare example of the failure of scientific consensus.
The point is that you can't know if an intervention works until you actually test it. There are tons of possibilities, including that some pm2.5 particles cause lung damage and some don't, in which case overall pm2.5 exposure would be correlated to negative health effects but using an air filter to reduce dust in your home might have little or no health benefit. We don't know yet.
Generally avoiding pm2.5 particles is a reasonable precaution but you can't say that things like air filters are actually the most effective health intervention someone can do until you test that experimentally. That's simply not how stuff like this works. There are virtually infinite health interventions that seem plausible but don't actually work when tested.
As with a treatment for alzheimer's disease, cancer, or heart disease, you can't simply say that X is correlated with disease, an intervention Y reduces X, so therefore Y must have positive health effects without actually testing Y.
It not the original title but I'm not sure it's "misleading"
> Within weeks, Binance fired or suspended at least four employees involved in the investigation, according to the documents and three people with knowledge of the situation. The company cited issues such as “violations of company protocol” related to the handling of client data.
That page is a few years old and it's much less the case now, which seems to disprove most of the broad cultural conclusions people are trying to draw based on it.
reply