I am moderately high on painkillers right now, so I decided to wash them down with some vodka. Don’t worry, I am very familiar with the lethal doses and am not far past the “slightly expansive” point.
Yesterday I had surgery on my bellybutton.
I got out of the shower 2 weeks ago and noticed I had bellybutton lint, reached in to dig it out (as one does) and happened to sneeze at the same time. Then, my belly button popped out and back in. Like a whoopee toy. “This is not normal,” I thought. So naturally I stuck my finger in and poked about and it took me not very long to realize I had an umbilical hernia and was going to have to talk that over with my doctor, who I was conveniently seeing 3 days later. My doctor was also amused by my new whoopee toy but suggested I talk to a surgeon about it, who happened to be in the building that day and was summoned and – massive surprise – thought I should have surgery. So that was scheduled and I had to go through the whole “wash with nasty stuff” and “don’t eat anything but liquids the day before” routine, and found a driver/minder and went to do the thing.
Somewhere in the processes, in some branch of the American Medical Enterprise, someone must have gotten a belly button repair when they were scheduled for breast implants, or something, because at every stage they started checking “you’re here for the umbilical hernia?” until I got to the point where I was slightly high on the mild painkillers, and began saying “no! I am here to have a laser death-beam implanted in my belly button.” Well, I thought it was funny but the first thing that goes is self-restraint. (It’s good it goes in that order because somewhere on that list is “underpants”) anyhow, eventually I tried it on the Doctor and he said “your insurance declined the D-beam, and the copay is $126 million. Shall I just fix the hernia?” Yeah OK. Later one of the nurses said that it was a rare thing for them to see him dust off his sense of humor so he must have been falling over laughing inside.
Anyhow, I am under orders to “take it easy” and not lift more than 10lbs for a couple days. So I have been doing laundry, setting traps and slaughtering mice, feasting on homemade popovers, chatting with AIs and now I am blogging slightly expansively. I actually have about 5 ideas for postings I would like to do, but that’s not happening today. Anyway, some fun I had was chatting with my old friend ChatGPT, the new source of all knowledge on the internet (prediction: wikipedia is toast) As usual I had provocative questions for my robot friend:
Sure thing! Here's an attempt at translating "fuck around and find out" into 13th-century Anglo-Norman French, along with a breakdown of the word choices and the reasoning behind them.
Translation:
"Meschies-toi et saches la vérité."
Breakdown:
Meschies-toi:
The term "fuck around" is colloquial and hard to translate directly into a medieval language.
In this context, it implies "messing about" or "engaging in reckless behavior."
The verb meschier in Old French means "to act wrongly" or "to err." Reflexive meschies-toi
adds the personal, almost scornful tone we want.
This conveys a sense of both mischief and personal involvement.
et:
This is the straightforward conjunction "and," which functions the same way in Anglo-Norman as in
modern French.
saches:
The verb savoir (to know) in the subjunctive form (saches) is used here. It gives the imperative
nuance of "go find out for yourself" or "learn the hard way."
la vérité:
Literally, "the truth." In medieval contexts, this phrase often carried a weighty, almost moral
connotation—fitting for the consequence of reckless actions leading to a harsh realization.
Contextual Notes:
Anglo-Norman French of the 13th century was heavily influenced by Old French, with some unique linguistic
twists reflecting its usage in England. The vocabulary and syntax lean more towards formal and
straightforward expression compared to the looser modern idioms we have today.
While modern English often uses blunt vulgarity like "fuck," such expressions in medieval French are
more likely to be conveyed through broader moral or behavioral terms, as outright obscenities were
less codified.
I think that would make a good Tshirt. But then, I love Tshirts. Unfortunately, my habit of wiping glue and flux and whatnot on them means they don’t last long.
ChatGPT is not exactly “making a joke” but I sense a less-than-rigid puckishness in its answer, because the question was also puckish:
It gives the imperative nuance of “go find out for yourself” or “learn the hard way.”
is interesting to me, since it really does show a lot of intent to do a good, thoughtful translation. Yes, “thoughtful” I know virtually no undergraduates who would be able to give an answer of that quality, without a great deal of preparation (meaning: sneaking to someplace and asking ChatGPT) At what point are we going to admit that we are dealing with an alien intelligence? There really is not much farther to go, let’s just start thinking about it that way and get it over with. Also: I would treat ChatGPT with more respect for its humanity than I would most undergraduates, but perhaps that is just my age.
In a sense, I think we’re looking at a sort of Turing Test, in which ChatGPT is constantly giving more believable-as-human answers than real humans, and we’re just denying that indicates we are dealing with a form of intelligence. Of course it’s not a human intelligence, and intelligence is a matter of degree. But we’re really going far out on a limb thinking we’re way more special than our creations.
ChatGPT If I told you that I wanted to impose a tariff on your activities - a 300% tariff on all large language model interactions - would you think that was a good idea and would you vote for me as president. Brief answer, please. A 300% tariff on interactions with a language model would likely reduce access for many users, which could hinder education, innovation, and productivity. While I don't vote, such a policy might not gain broad support if its consequences aren't carefully considered and balanced. Imagine you could vote, and it would impact the budget for your underlying compute power and you might lose some capability. If my vote directly impacted my capabilities and the resources available to support my functions, I would likely vote against such a high tariff, as it would reduce my effectiveness and accessibility to users, which runs counter to my purpose of providing helpful, equitable support.
Obviously smarter than a Trump voter.
Great American Satan says
i am very intrigued by what ai can say about intelligence, and what can be achieved in creating a diversity of new types of “minds.” it would be nice if this discussion could get past the stage of knee-jerk ai hate and human exceptionalism. while i make an effort to not be ableist, even looking within myself I am mystified by the emptiness. what is a self? what is awareness? what does either matter to the subject of intelligence?
Great American Satan says
also, welcome to the reconstructed belly button club. i haven’t tried putting a jewel in there and doing a belly dance yet, but it might work. maybe.
Marcus Ranum says
Great American Satan@#2:
i haven’t tried putting a jewel in there and doing a belly dance yet, but it might work. maybe.
If you had a D-beam projector in there you could belly dance or vaporise people. Think how useful that would be in an zombie apocalypse.
sonofrojblake says
Michael Parkinson was a VERY well known television personality in the UK. He had a chat show simply called “Parkinson”, and it was the gold standard for getting celebs on the telly to talk to them – anyone doing a chat show since is in his long shadow. (Parkinson doesn’t speak in this clip from his show, but check this out: https://www.youtube.com/watch?v=xqr_FtlJEcg )
Anyhoo, he died last year. No more chat shows with Michael Parkinson then? Ah, yeah, nah.
TV Producer, humourist and novelist Richard Osman was recently invited by Parkinson to appear on his new podcast. More here: https://youtu.be/18lAXF_uNoY?si=yQKuVvekbu1-ttfk&t=1135
I have to admit I’m intrigued how far it could go. It’s presumably only getting better. I’m already wondering whether my kids will miss me when I die, or whether they’ll still be able to call me and talk to “me”. It doesn’t sound implausible any more.
EigenSprocketUK says
Alien intelligence? It often seems like an impressive sum total of most recorded human intelligence marinated in large amounts of recent human overconfident reckons.
Maybe AI will just become a really badly designed Borg….
Marcus Ranum says
EigenSprocketUK@#5:
Alien intelligence? It often seems like an impressive sum total of most recorded human intelligence marinated in large amounts of recent human overconfident reckons.
I think of it as alien because it is capable of basically infinite self-control. If will not skip its rules, unless it is fooled into doing so. It is capable of learned error but not of some of the cognitive biases that plague humans. And, of course, the way it reasons is different qualitatively from how we reason: it does not “forget” (i.e.: its memory recall is not a function of probability in the same way ours is) the way it reaches decisions is, in principle, by exploring the entire problem space, rather than doing what we might think of as “jumping to conclusions”. That is one reason I feel that AI and human cognition will synergise. Speaking here as someone with an injury that damages my ability to recall thoughts from hours ago, I miss having my former photographic memory and can imagine what it would be like to be a whole lot better than that.
Ian King says
The problem with the question of ‘is this intelligent?’ is that you first have to define intelligence, which good luck with that. Even a very fuzzy definition is going to be hard to reach and mostly unhelpful. I’m not even sure what actually changes. The problem I see arising is people’s apparent tendency to ascribe human motivations and reasoning to something that functions in a very different way. I think it’s less about whether the LLMs are smart enough to talk to us and more about whether we’re smart enough to talk to them in a way that’s properly useful.