Michael Roe

Karma: 114

Michael Roe 7 Jun 2024 9:30 UTC
1 point
0
in reply to: Michael Roe’s comment on: Is Claude a mystic?
The poor naive viewer thought Neon Genesis Evangelion was just going to be about giant robots fighting, but, oh no, there seems to be a Religious Allusion.

Michael Roe 7 Jun 2024 9:27 UTC
3 points
0
on: Is Claude a mystic?
The Quest for the Holy Grail is, in origin at least, religious.
So, maybe, the LLM is picking up on a textual similarity between (a) religious allegory; (b) Dungeons and Dragons type adventures that retain some of the trappings of religious allegory.
Could be a type of Waluigi effect: once you’ve started making it a religious allegory, you’re stuck continuing with it, and at each stage there is some chance that your amoral and nihilistic adventurers (in th efashion of Fritz Lieber’s Lankmar) will suddenly discover Spiritual Signifcance.

Michael Roe 2 Jun 2024 21:06 UTC
5 points
2
on: How do you shut down an escaped model?
Serious answer: government already makes requests to e.g. DNS hosting providers to shut things down.
Non-serious answer: Turing Police, like in the William Gibson novels.

Michael Roe 1 Jun 2024 18:10 UTC
2 points
0
in reply to: Mascal's Pugging’s comment on: Awakening
I’ve lost count of how many gurus have been involved in some sort of sex scandal...
(Seriously, care would seem to be advisable when choosing a yoga sex cult.)

Michael Roe 1 Jun 2024 11:41 UTC
7 points
0
on: Awakening
Thanks for this post. I think that these kind of bad experience reports are valuable.
1. Is meditation-induced psychosis really a thing? Sure looks like at this point, as we have a significant nuber of well-documented case histories.
2. Well. how risky is meditation then? I am currently unsure on this point; the sheer number of bad experience reports, is somewhat concerning. By way of comparison: I am currently taking carbimazole for Graves disease, a drug with a maybe 1E-3 (ish) probability of a side effect that is kind of bad (agranulocytosis) … well, I might be willing to take a 1 in 1000 risk for the benefit of the drug.
3. Were there particular risk factors in these case reports, that indicate meditation is contra-indicated for some type of people? Also, not currently clear to me.

Michael Roe 1 Jun 2024 11:29 UTC
1 point
0
in reply to: Tapatakt’s comment on: Awakening
In Buddhist philosophy, there’s a type of argument along the lines of … suppose that you actually attained enlightenment. yeah, sure, sounds unlikely, as enlightenment seems kind of hard. but .. just hypothetically supposing you did … and given what we believe “enlightenment” is supposed to be like. Then, would that “you” still be “you”?

Michael Roe 5 May 2024 10:48 UTC
8 points
−1
on: Some Experiments I’d Like Someone To Try With An Amnestic
This sounds like a terrible idea.
Though, if you’re going to be put under sedation in hospital for some legit medical reason, you could have in mind a cool experiment to try when you’re coming around in the recovery room.
i was sedated for endoscopy about 10 years ago,
they tell you not to drive afterwards (really, don’t try and drive afterwards)
and to have a friend with you for the rest of the day to look after you
i was somewhat impaired for the rest of the day (like, even trying to cook a meal was difficult and potentially risky … e.g. be careful not to accidentally burn yourself when cooking)
I drew a bunch of sketches after coming round to see how it affected my ability to draw.

Michael Roe 25 Apr 2024 12:19 UTC
3 points
−2
on: Is being a trans woman (or just low-T) +20 IQ?
Alternative theory (which, to be clear, I dont actually believe, but offer for consideration)
- Many of the high iq people are too autistic to be successful
- but female hormones protects against the autism somehow, without impacting iq too much
- so the successful high iq people tend to be trans more often on average

Michael Roe 25 Apr 2024 12:05 UTC
12 points
11
on: Is being a trans woman (or just low-T) +20 IQ?
I think its more likely its the transgender—autism correlation....
- some forms of autism come with higher iq (and other forms, really really dont)
- and there’s the transgender autism correlation
which together would seem to predict transgender high iq people
(and also transgender low iq that you arent seeing due to ascertainment bias)

Michael Roe 18 Apr 2024 20:24 UTC
4 points
2
in reply to: Michael Roe’s comment on: Paul Christiano named as US AI Safety Institute Head of AI Safety
And the really funny bit is NIST deliberately subverted the standard so that an organization who knew the master key (probably NSA) could break the security of the system. And then, in actualt implementation, the master key was changed so that someone else could break into everyone’s system And, officially at least, we have no idea who that someone is. Probably Chinese government. Could be organized crime, though probably unlikely.
The movie Sneakers had this as its plots years ago.. US government puts a secret backdoor in everyone’s computer system .. and, then, uh,, someone steals the key to that backdoor;
But anyway, yes, it is absolutely NISTs fault that they unintentionally gave the Chinese government backdoor access into US government computers.

Michael Roe 18 Apr 2024 20:15 UTC
3 points
0
in reply to: habryka’s comment on: Paul Christiano named as US AI Safety Institute Head of AI Safety
https://cacm.acm.org/research/technical-perspective-backdoor-engineering/
for example. Although that paper is more about, “Given that NIST has deliberately subverted the standard, how did actual products also get subverted to exploit the weakness that NIST introduced.”

Michael Roe 18 Apr 2024 12:57 UTC
1 point
0
in reply to: Michael Roe’s comment on: lcmgcd’s Shortform
While I was typing this, quetzal_rainbow made the same point

Michael Roe 18 Apr 2024 12:53 UTC
1 point
−1
in reply to: Michael Roe’s comment on: lcmgcd’s Shortform
Ascertainment bias, of course, because we only see the cases where this did not work, and do not know exactly how many members of e.g. Delta Force were originally in doubt as to their gender. We can know it doesnt work sometimes.

Michael Roe 18 Apr 2024 12:41 UTC
1 point
0
in reply to: lukehmiles’s comment on: lcmgcd’s Shortform
Well there’s this frequently observed phenomenon where someone feels insecure about their gender, and then does something hypermasculine like joining Special Forces or becoming a cage fighter or something like that. They are hoping that it will make them feel confident of their birth-certificate-sex. Then they discover that nope, this does not work and they are still trans.
People should be aware that there are copious examples of people who are like—nope, still trans—after hoping that going hard on their birth-certificate-gender will work,

Michael Roe 18 Apr 2024 12:27 UTC
8 points
0
on: Paul Christiano named as US AI Safety Institute Head of AI Safety
Ever since NIST put a backdoor in Dual Elliptic Curve Deterministic Random Bit Generator, they have the problem that many people no longer trust them.
I guess it might be possible to backdoor AI Safety Evaluations (e.g. suppose there is some know very dangerous thing that National Security Agency is doing with AI, and NIST deliberately rigs their criteria to not stop this very dangerous thing).
But apart from the total loss of public trust in them as an institution, NIST has done ground-breaking work in the computer security field in the past, so it wouldn’t be so unusual for them to develop AI criteria.
The whole dual elliptic curve fiasco is possibly a lesson that criteria should be developed by international groups, because a single country’s standards body, like NIST, can be subverted by their spy agencies.

Michael Roe 15 Apr 2024 19:29 UTC
3 points
0
on: Reconsider the anti-cavity bacteria if you are Asian
A Japanese guy I used to work with had a very serious genetic alcohol intolerance. (Like, a single drop of wine would probably be ok but anything more than that would likely put him in the hospital).
I guess if you have a known inability to metabolise alcohol, such that you’re already having be very car3fuo that anything you consume doesn’t have alcohol in it, you might want to be a little bit cautious here. but … maybe the quantity you get from the bacteria is so small it doesn’t matter.

Michael Roe 18 Mar 2024 22:33 UTC
11 points
3
on: The Worst Form Of Government (Except For Everything Else We’ve Tried)
The Northen Ireland Assembly works this way, at least for some things.
But, in general, the U.K. does not work that way. A particular political party sometimes gets a big majority.

Michael Roe 14 Mar 2024 17:38 UTC
3 points
0
in reply to: Samuel Hapák’s comment on: The Parable Of The Fallen Pendulum—Part 1
This depends very much on how well debugged the compiler is...
* gcc on llvm on Intel hardware … very unlikely to be a bug in the compiler
- you’re on some less well exercized target like RISC-V … ha, you are in for so much pain
it is so much fun debugging on experimental hardware where any of (a) your program (b) the compiler (c) the actual hardware are all plausibly buggy.
oh, I forgot (d) the tool used to convert the hardware description language (used to specify the chip design ) into logic gates, used to build the hardware, is itself buggy

Michael Roe 11 Mar 2024 19:35 UTC
2 points
0
on: Some (problematic) aesthetics of what constitutes good work in academia
As someone who has worked in both academia and industrial research labs, in both cases you can claim either academic publcation or real-world impact as a success wrt getting promoted …
a) I got a paper about this published in a top-ranking journal; vs,
b) look, those guys are now selling a product based on this thing I invented
(in an industrial research lab, “those guys” had better be the product division of your company; if you’re an academic funded by DARPA, “those guys” being anyone who is paying taxes to the US government is just great)

Michael Roe 6 Mar 2024 15:54 UTC
4 points
1
on: My Clients, The Liars
(With a nod to SBF). What kind of criminal mastermind creates a Signal chat group called “wirefraud”? are you going to try and tell me there was a perfectly innocent explanation?