benmmurphy
3d ago
If this was true I have no idea how this didn't get him killed. There seems to be two outcomes. You go to jail, or someone is going to flip out because you didn't go to jail and murder you.
benmmurphy
10d ago
Photos remind me of the Capitol from the hunger games.
benmmurphy
11d ago
Mass AI cheating would fix the achievement gap and make it so the students who have fallen behind don't look like they have fallen behind. Ubiquitous AI cheating is potentially a massive gift for schools and universities. I guess with universities there is a risk it might destroy the reputation of the university. but this is a problem someone else will have to deal with in 5 years time. The current administrators are free to set fire to the schools reputation and enjoy all the rewards that come with it.
benmmurphy
25d ago
The banking system is already an investigative part of law enforcement. It would be just another crime to add to the list of crimes they are responsible for investigating. I'm not arguing having the banks perform this role is a good idea but that ship has already sailed.
benmmurphy
26d ago
Hanania dropping the sarcasm in the twitter thread:
I know right! Lmao, just like they told us to take the vax, fellow pureblood.
benmmurphy
1mo ago
I have two datapoints about AI and programming recently.
-
I asked it about an unknown PRNG function I've reverse engineered which I had previously tried googling to see if it was based on a standard function. It was able to find functions that were similar that I had not previously been able to find googling. I then asked it to come up with a known plaintext attack when part of the seed was known and it spat out something that looked correct.
-
Another developer was looking at reverse engineering a function that was protected with a weak form of control flow obfuscation. The control flow obfuscation was just replacing function call instructions with a call to a shared global dispatch function that would end up calling the target function. The global dispatch function would execute approximately 200 instructions. There is an obvious attack against this obfuscation and it can be stripped off with ~100 lines of python in ghidra. They were using LLMs to try and investigate this function but didn't make much progress. But maybe with better prompting and allowing more access to tools it would have been possible for the LLM to make progress.
benmmurphy
1mo ago
isn't a 10 year ban better than a bill which just bans states from regulating AI. at least the 10 years creates is a sunset clause on the regulation and would require congress to pass new legislation if they think continuing the ban is a good idea. though, maybe generally we should be pushing congress to include short sunset clauses in all legislation it passes because the future could be very different in X years.
benmmurphy
1mo ago
right wing housing theorem of theory sounds a bit like high housing prices suppress TFR and this leads to an increase in immigration in order to maintain high housing prices. not sure if the data is consistent with that. i guess left wing housing theory of everything wouldn't include immigration but include inequality and some other left wing focused issues.
benmmurphy
1mo ago
it might hurt greenwald's reach with the normies. whenever someone brings up glen's reporting with normies someone else can point to the sex videos to derail the conversation.
benmmurphy
1mo ago
its might be a good thing. at the moment there is some value from pushing false information but if there is monetary value from generating false information then hopefully this will end up pushing the value and monetary value from pushing false information close to zero. there is some kind of commons that these false information spreaders are farming but once the barriers are removed and there are monetary incentives the commons is going to be destroyed.
benmmurphy
1mo ago
isn't that just the meme about questions at academic lectures. its not usually about asking a question, its usually just the person pushing their hobby horse.
benmmurphy
1mo ago
presumably, you can just compare deaths across a covid and non-covid period to get a rough estimate of covid deaths. i doubt the policies put in place to fight covid led to a large number of extra deaths in the short term.
benmmurphy
1mo ago
The enforcement/investigation for KYC/AML looks like a 4th amendment violation or at least looks like its structured to do something that would be a 4th amendment violation if the government directly did the thing.
benmmurphy
2mo ago
I suspect most countries now have some form of anti-trust legislation. Wikipedia has some details on the price fixing page: https://en.wikipedia.org/wiki/Price_fixing However, there may have been periods of time where countries had strong unions but no anti-cartel legislation. I think Australia only cracked down on price fixing after 1974.
benmmurphy
2mo ago
didn't Michael Jackson do something similar with 'they don't care about us'. i was shocked when i first heard the original words to that song and they were used as background music for a runners instagram.
benmmurphy
2mo ago
People keep complaining that people are worried about AI causing human extinction are not carrying out terrorism. But maybe they already have. There is a conspiracy theory that Malaysia Airlines Flight 370 crashed because a government wanted to kill Freescale Semiconductor staff. But maybe the Butlerian Jihad wanted to kill Freescale Semiconductor staff.
benmmurphy
2mo ago
the claim that capital controls are ok for dealing with imbalances but tariffs are bad seems questionable. if they both end up making changes in trade to change the imbalance then it would seem both would have the same deadweight losses associated them. i can understand that maybe capital controls have some large side benefit that tariffs do not which would offset the deadweight loss from the change in trade. for example God could come down from heaven and because he supports capital controls he could dump container loads of semi-conductor chips into the US like mana from heaven. this would be a massive benefit that would not normally appear if you implemented tariffs. but i strongly suspect there are no special large benefits from capital controls that would offset the same deadweight loss tariffs also suffer from.
the classical argument against tariffs is the change in trade they produce is bad. so if you find some technique that produces the same change then it just seems like you have come up with some abstraction to try and hide that you are doing the bad thing. tariffs would be a great tax if they did not produce a change in trade.
Sounds inspired from Warhammer 40k or Warhammer 40k was inspired by this mess. "Imperium" of man, 9 loyal legions and the Warp corresponding to the 'acausal realm'. Maybe the 1980s D&D moral panic was onto something.
benmmurphy
3mo ago
its like a reverse of the Chinese generals meme.
What is the punishment for being in America illegally?
Being deported from America
I better leave America voluntarily then.
benmmurphy
3mo ago
If you check the raw emails you can also see someone warned them about controlling for birth weight but they ignored the warning.
https://cdn01.dailycaller.com/wp-content/uploads/2025/03/U-MN-FOIA-concordance.pdf
I was talking with a health economist about it & he asked 2 questions.
...
[2] When discussing possible mechanisms, he asked whether there is an effect on birth weight (and I would add gestational age). I said that this is conditional on the hospital stay when giving birth, so there's not much scope for affecting these at that point. He suspected that the patient-physician relationships might pre-date birth and we might see a legitimate good effect on it. Alternatively, seeing nothing could be interpreted as evidence against selection, a placebo. These competing interpretations make it a bit dicey.
[2] We do have controls for low weight and early term in the estimations. Do you want me to split based on these?
[2] No. The idea would be to use them as outcomes, rather than controls. But I'm not sure we want to go down that road. It could be a mechanism but it requires that the baby's doctor (not the mom's) exert some influence during the pregnancy. Given that we think the mother's doctor is more important prenatally than the baby's, it would be more informative if we could see the effect of mother's doctor race on child birth weight and gestation. But this is not possible, I think.
benmmurphy
3mo ago
In sports i think the 4 minute mile barrier is a good example. This was broken in 1959 by Roger Bannister who had relative poor training even at the time and still has not been broken by a woman with the fastest women's time being ~4:07. While since 1959 the men's record has dropped to 3:43. 24 seconds over this distance feels like an insane difference. This came to me this morning because I was watching some younger women's US national mile event and the national record was 4:27 but I suspect a lot of amateur club runners would be able to break this time. Also, if you ask ChatGPT about differences between men and women in sports it will say there are physical difference but overall give you a non-answer. So if you believe ChatGPT is some kind of summary of human knowledge then that could explain how someone could have the wrong idea if they are summarizing the same way ChatGPT does.
benmmurphy
3mo ago
I was also banned from going to the US but no one would wrote up my sob story, but I also didn't break the law by entering the US so no real story to write up.
benmmurphy
3mo ago
otherwise everyone coming to the US as a conference attendee wouldn't be able to use the visa waiver program
benmmurphy
3mo ago
allegedly the hacker has some kind of anime meme account on twitter: https://x.com/jeremykauffman/status/1903559150023344354
The NYU hacker working to save western civilization appears to be a young gentleman called @bestniggy who follows me For most of my life that sentence would have surprised me but now it seems almost normal
The problems of LLMs and prompt injection when the LLM has access to sensitive data seem quite serious. This blog post illustrates the problem when hooking up the LLM to a production database which does seem a bit crazy: https://www.generalanalysis.com/blog/supabase-mcp-blog
There are some good comments on hackernews about the problem especially from saurik: https://news.ycombinator.com/item?id=44503862
The problem seems to be if you give the LLM readonly access to some data and there is untrusted input in this data then the LLM can be tricked into exfiltrating the data. If the LLM has write access to the data then it can also be tricked into modifying the data as well.
More options
Context Copy link