Artificial intelligence officially gets scary

[woops]

Actually I’m thinking about reading Faust next, as I’m unfamiliar with the story. I’ve seen the film though, just don’t remember most of it.

oi stan save yourself a lot of work by reading marlowe's faust, not the mega long goethe version, and enjoy some wicked elizabethan style "methinks he has become mad / by being ouer solitary."

 

[Clinamenic]

But I think we shouldn’t entirely narrow in on the dystopian aspects of all this. We have a virtually infinite library of premier research, discourse, art and literature at our fingertips, for free. We have payment systems available that are 100% transparent. These things resulted within a neoliberal capitalist paradigm, even as unprecedented wealth was amassed by a tiny subset of humans. To focus on the dark side is to unnecessarily sentence yourself to depression and alienation.

 

[Clinamenic]

oi stan save yourself a lot of work by reading marlowe's faust, not the mega long goethe version, and enjoy some wicked elizabethan style "methinks he has become mad / by being ouer solitary."

I already have the Goethe version, and was also just interesting in reading something of his, but I didn’t know there was another version. Is it that different?

 

[Leo]

This sort of thing is why I find Gus' tone unsettling at times, the way he favours this very mechanical vision of things like human interactions and breaks them down into optimal strategies.

also, the trust in human nature. positive scenarios could play out if man was inherently good, on a mission to attain what's best for mankind. we know from history that isn't always, of even frequently, the case.

 

[Clinamenic]

But I think we shouldn’t entirely narrow in on the dystopian aspects of all this. We have a virtually infinite library of premier research, discourse, art and literature at our fingertips, for free. We have payment systems available that are 100% transparent. These things resulted within a neoliberal capitalist paradigm, even as unprecedented wealth was amassed by a tiny subset of humans. To focus on the dark side is to unnecessarily sentence yourself to depression and alienation.

Whereas to narrow in on the bright side is to become less vigilant regarding any threats to civil liberties and luxuries.

 

[Mr. Tea]

Fascinating article in the New Yorker about fMRI scans tracing iron atoms involved in neuronal oxygenation, and mapping this activity onto the stimuli being experienced by the patient, like watching a Hitchock movie or hearing a series of words. A means of programmatically mapping out thought space. Not much to do with AI, although of course machine learning was involved in analyzing these vast data sets (square millimeters or "voxels" of brain matter, presumably grey matter but the article didn't specify). Naturally, iARPA was involved:

The work at Princeton was funded by iARPA, an R. & D. organization that’s run by the Office of the Director of National Intelligence. Brandon Minnery, the iARPA project manager for the Knowledge Representation in Neural Systems program at the time, told me that he had some applications in mind. If you knew how knowledge was represented in the brain, you might be able to distinguish between novice and expert intelligence agents. You might learn how to teach languages more effectively by seeing how closely a student’s mental representation of a word matches that of a native speaker. Minnery’s most fanciful idea— “Never an official focus of the program,” he said—was to change how databases are indexed. Instead of labelling items by hand, you could show an item to someone sitting in an fMRI scanner—the person’s brain state could be the label. Later, to query the database, someone else could sit in the scanner and simply think of whatever she wanted. The software could compare the searcher’s brain state with the indexer’s. It would be the ultimate solution to the vocabulary problem.

Jack Gallant, a professor at Berkeley who has used thought decoding to reconstruct video montages from brain scans—as you watch a video in the scanner, the system pulls up frames from similar YouTube clips, based only on your voxel patterns—suggested that one group of people interested in decoding were Silicon Valley investors. “A future technology would be a portable hat—like a thinking hat,” he said. He imagined a company paying people thirty thousand dollars a year to wear the thinking hat, along with video-recording eyeglasses and other sensors, allowing the system to record everything they see, hear, and think, ultimately creating an exhaustive inventory of the mind. Wearing the thinking hat, you could ask your computer a question just by imagining the words. Instantaneous translation might be possible. In theory, a pair of wearers could skip language altogether, conversing directly, mind to mind. Perhaps we could even communicate across species. Among the challenges the designers of such a system would face, of course, is the fact that today’s fMRI machines can weigh more than twenty thousand pounds. There are efforts under way to make powerful miniature imaging devices, using lasers, ultrasound, or even microwaves. “It’s going to require some sort of punctuated-equilibrium technology revolution,” Gallant said. Still, the conceptual foundation, which goes back to the nineteen-fifties, has been laid.

edit: typo fixes in the first sentence, from "experience" to "experienced" and from "word" to "words"

Combine this tech with VR and deepfakes, and we could be standing in the brink of a pornographic utopia the likes of which would have boggled the mind even of David Foster Wallace.

 

[Clinamenic]

also, the trust in human nature. positive scenarios could play out if man was inherently good, on a mission to attain what's best for mankind. we know from history that isn't always, of even frequently, the case.

Yes, and this plays out interestingly when we are designing systems that can take into account, and even guide, people’s incentives. To what extent should these systems assume the worst in people, I.e assume that they will defect at any opportunity that seems beneficial to them? And to what extent does making this assumption actually reify it, I.e to what extent does assuming the worst people bring out the worst in people?

I do have faith in nature, and that include human nature.

 

[woops]

I already have the Goethe version, and was also just interesting in reading something of his, but I didn’t know there was another version. Is it that different?

i'm pretty sure it's the same story but told in a short play rather than 3 massive volumes or whatever the goethe is and probably a lot more fun

 

[sus]

This sort of thing is why I find Gus' tone unsettling at times, the way he favours this very mechanical vision of things like human interactions and breaks them down into optimal strategies.

For whatever it's worth, I don't find my world disenchanted as a result. It's the old thing about the flower—a botanist is gonna probably have an experience of more awe and appreciation than your average viewer. "Oh, it's just a normal old white flower like a daisy." "No, look at the subtle way the stamen curves, very irregular, but so elegant a solution... And this coloration, you can only get this coloration if the ground is high in XYZ mineral content, so there must be a lot of iron in the soil... And look, it's chosen to only make a single flower, perhaps because of the lack of water."

Nature's complexity is so limitless and incredible and breathtaking that you nearly always get more out of the thing the more you know, rather than less. This view that knowledge destroys the mystery, and turns nature into this simple mechanical process, stripped of all meaning, just isn't true IMO. And the same goes for language, social life, etc

 

[Clinamenic]

Combine this tech with VR and deepfakes, and we could be standing in the brink of a pornographic utopia the likes of which would have boggled the mind even of David Foster Wallace.

Yeah no need for deep state pedo rings if it can all be simulated!

 

[Leo]

But I think we shouldn’t entirely narrow in on the dystopian aspects of all this. We have a virtually infinite library of premier research, discourse, art and literature at our fingertips, for free. We have payment systems available that are 100% transparent. These things resulted within a neoliberal capitalist paradigm, even as unprecedented wealth was amassed by a tiny subset of humans. To focus on the dark side is to unnecessarily sentence yourself to depression and alienation.

True, and as a natural/jaded sceptic, I have to sometimes check myself to make sure I'm weighing the good against the bad. I think what sets me off here is the cold detachment-bordering-on-nonchalance of how the future plays out.

hey, gotta break some eggs to make an omelette, right?

 

[Mr. Tea]

Yeah no need for deep state pedo rings if it can all be simulated!

Well there's still the question of the adrenochrome harvesting, of course.

 

[sus]

“The mysteries of mimicry had a special attraction for me,” Vladimir Nabokov writes in “Butterflies.” Nabokov emphasizes that the imitative practices of the butterflies take on a special aesthetic wonder because their reason is so unknown, so enigmatic and “nonutilitarian.” Natural selection, Nabokov argues, “in the Darwinian sense, could not explain the miraculous coincidence of imitative aspect and imitative behavior, nor could one appeal to the theory of the ‘struggle for life’ when a protective device was carried to a point of mimetic subtlety…[it was] a form of magic… a game of intricate enchantment and deception." This approach is a common one; Burke, as chronicled in Butterfly People, believed that nature lost its “magic and charm” as her secrets were unveiled.

(There was, of course, a Darwinian explanation for butterflies’ mimicry of nature hypothesized by Henry Bates as early as 1866, of which Nabokov was merely ignorant. “Palatable” species — those which could be safely consumed or had nutritional value for predators — would take on the patterns of “unpalatable” species, surviving because they would not be preyed on.)

Despite the demystification of mimicry — alongside many other natural phenomena — contra Burke, the natural beauty and aesthetic wonder of the butterfly appears to have been far from diminished in the eyes of American lepidopterists. To many naturalists of the era, scientific understanding and aesthetic appreciation went hand in hand, each supplementing the other. The artistry and visual beauty of animals’ outer appearance, described in Animal Forms and Patterns by Adolf Portmann, is derived from an understanding of its relation to the animal’s inner form and entirety of function. The more one visually appreciates (or “intensively absorbs”) a living thing, the more “unsuspected delights” are disclosed to the eye, while simultaneously “deeper insight into [the forms’] significance” is revealed. Again, “viewing” is distinguished from truly seeing, the latter leading to increased aesthetic pleasure due to an understanding of a species or form’s “how” and “why.” Without an understanding of that why, Portmann or Humboldt might argue, Nabokov is shorted: he may have sported keen observations of butterfly mimicry, beautifully recorded and described, but without an appreciation of the Darwinian “why,” Nabokov is doing something short of truly seeing.

Thus, the twin motivations of collecting for aesthetic reasons, and collecting for scientific knowledge, were inextricably intertwined. Though the lepidopterists documented in Butterfly People “began their careers… awash in the heat and smells of the meadow and forests, sensitive to something worth losing oneself in” they sought "unceasingly to know. Understanding helped make “visible the invisible," revealed the depths of its beauty and character. Knowledge replaced sensory ecstasy much like the maturity of adulthood replaces the passions of childhood.

 

[Clinamenic]

True, and as a natural/jaded sceptic, I have to sometimes check myself to make sure I'm weighing the good against the bad. I think what sets me off here is the cold detachment-bordering-on-nonchalance of how the future plays out.

hey, gotta break some eggs to make an omelette, right?

Yeah admittedly I do come off as detached, and depending on how you define it, I am. Could just be another way of articulating the enlightenment of “seeing through it all”. As I understand it, detachment is something that is both positive and negative in different respects. Allegedly Himmler carried around a copy of the Bhagavad Gita.

 

[Clinamenic]

I wasn't talking about disenchantment in that instance, I was talking about a clinical approach to things like speaking to people and viewing them as machines to be tweaked and tuned.

But I would argue that, at certain levels of scientific understanding, more and more aspects of a person do become tweakable and tunable, or at least so thoroughly understandable.

 

[sus]

I wasn't talking about disenchantment in that instance, I was talking about a clinical approach to things like speaking to people and viewing them as machines to be tweaked and tuned.

It's not that that's how communication feels from the inside. It's that, from an external perspective, it's what is revealed by the patterns of action and response of social encounters. When a boy bird puffs up his feathers around a cute girl bird, he might do it because it feels good, but we know the evolutionary reason. It's the same with human communication. I guess the question is, does knowing the rationale from this external, objective, evolutionary perspective taint the phenomenology, make you feel like a robot manipulator, and in my experience it doesn't. YMMV tho

 

[Leo]

Allegedly Himmler carried around a copy of the Bhagavad Gita.

the prosecution rests, your honor.

 

[sus]

I will say, we are so so so much worse at this right now than people think

For all the talk of social media manipulation and nudges, behavioral economics and gamification theories are so crude, there are major replication crises of studies, most of the touted "outsized" effects (where minor interventions yield large behavioral effects) are basically bullshit.

ML systems that learn through experimentation will probably become pretty good though, and likely are already outperforming our academic theories.

What I think we'll end up with is a place where we know, given a certain context, based on ML training, what works. But we won't know why. We won't have grand theories, just a set of super contextually specific interventions that are effective.

 

[Clinamenic]

Yeah, and I'm saying that's one of the dangers of our current trajectory. It seems as though it's inevitably going to lead to some people being able to tweak the rest of us to their own ends and to a much greater degree than they already do.

i think it will be a mix of us being spontaneously molded by circumstances beyond the control of even the elite, as well as by circumstances within their control and according to their will. I don’t think we’ll reach the point where humans dictate the fate of humanity, nor do I think the popular will is to be slept on.

That said, I’d agree that the power elite entities exert over human evolution, culturally and genetically, will continue to reach unprecedented heights, but so will the degree of natural complexity in which these efforts are contextualized.

 

[sus]

Think how crude advertising still is—how many products have you purchased that you've come across online? A handful at most, I'd wager. Out of tens or hundreds of thousands. Incredibly crude stuff.