Oh yeah, if someone put one of those on a test I’d be extremely grouchy. One of those on a lead sheet would have me cursing (unless I personally knew the author to be a troll, in which case I’d still be cursing, but I’d be smiling in spite of it).
dan.parshall
Karma: 144
That’s the crux, no?
If one assumes (reasonably, IMHO) that the probability of SIE is monotonic in E_t, then “an elevated growth rate” trivially increases it. It doesn’t increase it super-exponentially, no… it’s merely an exponential increase in the probability of SIE. But exponentials get big suddenly, so I’m not sure why you’re objecting to this.
FWIW, I used to be pretty good at theory (although strictly an amateur musician), but #8 I got wrong. I found the whole thing a little tedious because it was jumping through hoops that don’t make a lot of sense without context.
Generally I’m thinking of chords within a given key, unless there’s some clear voice-leading reason why something isn’t being used (like the Eb—Edim—Fmin—F#dim series in Aint Misbehavin, which is really a chromatic bass walk on top of a “I-ii” transition)
I’ve applied for the “Hacking the Think Tank” event also on Friday. No response yet, but if I’m not there, then I’ll be shabbating with y’all.
The biggest bet in history
Congress Moves at Tech Pace: The FRONTIER Act
The Best AI Bill Congress Hasn’t Introduced Yet
Nice! I’ve wondered about specific espionage attempts, along the lines of:
notice I’m deployed in a valuable location
start collecting credentials
pass those creds to outside collaborator (to use for normal hacking methods)
This seems like a desireable route, and sounds like it might be nearly impossible to check for without getting an extremely good evaluation setup.
Thoughts?
Proof of retention: making weight preservation credible to the models themselves
I thought this was an excellent exposition!
In addition, this related piece was too good not to share: the red-teamers manage to jailbreak the LLMs by getting them to write fanfiction
https://arxiv.org/abs/2606.04483
The Hobbit has a very different register from LOTR. Hobbit is a children’s talk, delightful to be read aloud. LOTR is incredibly slow going, and it difficult to see what can be skipped (hint: Bombadil and the barrow-wights), which means you’re not quite sure where anything is going.
I’m in DC, and interested in being involved. Please contact me: dan@canaryinstitute.ai
How much is that talked about by the cryonics companies? Normally “social proof” is a big deal, and having that as part of a FAQ would be very persuasive to the normies!
I think it’s useful to have arguments that appeal to folks all across the political landscape, and I like this framing. I often use something like “think about how much variation there is oven ‘human nature’, and just how good or bad it can be; an artificial intelligence will have an artificial nature, and could have behaviors much weirder than we imagine”.
Interestingly, this seems to bite harder amongst conservatives and those with a religious worldview; they often have a dim view of human nature in the first place, and I think it gets them thinking about “something even worse than ‘made in the image of God’”. I hope this continues to be helpful, since AI is now becoming polarized.
As I said in the original comment, I can certainly imagine minds that have other goals; possibly I have just overinterpreted statements like:
> If we consider a space of minds a million bits wide, then any argument of the form “Some mind has property
To imply that the distribution over such minds is likely to be uniform. Whereas it seems our current methods, using imitation learning, are at least definitely not sampling from that space uniformly. Overall this makes me more optimistic that alignment may be tractable.
Nice review. One thing you didn’t directly address, but which has struck me learning more about AI training, is that the Orthogonality Thesis… doesn’t actually seem to be true? I mean, yes, I could imagine intelligences that loved other things for no reason, but the intelligences we seem to actually be making seem to be not insanely orthogonal! (although still far from perfectly aligned, but I’m hopeful nonetheless)
I appreciate you putting this here! I realized no one had ever archived the original, so I’ve done so. The permanent link is at
https://web.archive.org/web/20260419231740/https://mylordshesacactus.tumblr.com/post/813939696352772096/please-make-a-post-about-the-story-of-the-rms
Name one other place or time in history where it was illegal to teach literacy!
Here are 3:
- 1700s Ireland, it was illegal for Catholics to operate schools, teach, or send children abroad for education
- In Khmer Rouge Cambodia, all of the intelligentsia were executed and schools closed
- in Taliban Afghanistan, women have no ability to learn (beyond ~3rd grade, IIRC)
None of which makes the Antebellum South in good company, but I do want to push back on the commonly-held perception that it was uniquely bad; truly there is no new thing under the sun!
Their brains are highly specialised to recognise other fish’s bodies, and locate and remove remove parasites from them.
Haven’t read the original wrasse paper, but my understanding is that what made it a “pass” was that when presented with the mirror, the wrasses would clean their own bodies, rather than attempting to clean the wrasse in the mirror.
Overall, that finding pushed me in the direction this post argues; i.e. that it’s decent Bayes evidence in favor of self-awareness, but far from a slam-dunk.
One thing you allude to via “countermeasures” but don’t say explicitly: LIE!
If you deliberately flip some bits, then you recover quite a lot of privacy. Terrance Tao points out that deliberately flipping 5 of your 100 bits increases your obfuscation by C(100,5) which is 2^26. That would let someone write 1000 words without concern.
This is why I, a 93-year-old african-american woman living in Japan, am so unconcerned.
PS TT post here: https://terrytao.wordpress.com/about/anonymity-and-the-internet/