Friday, March 31, 2023
Are we publishing too damn much?
There's too much science: “These findings suggest troubling implications… If too many papers are published in short order, new ideas cannot be carefully considered against old & processes of cumulative advantage cannot work to select valuable innovations” https://t.co/os1xxwTCKl pic.twitter.com/vZbCgTDEN3
— Ethan Mollick (@emollick) March 31, 2023
Significance:
The size of scientific fields may impede the rise of new ideas. Examining 1.8 billion citations among 90 million papers across 241 subjects, we find a deluge of papers does not lead to turnover of central ideas in a field, but rather to ossification of canon. Scholars in fields where many papers are published annually face difficulty getting published, read, and cited unless their work references already widely cited articles. New papers containing potentially important contributions cannot garner field-wide attention through gradual processes of diffusion. These findings suggest fundamental progress may be stymied if quantitative growth of scientific endeavors—in number of scientists, institutes, and papers—is not balanced by structures fostering disruptive scholarship and focusing attention on novel ideas.
Abstract:
In many academic fields, the number of papers published each year has increased significantly over time. Policy measures aim to increase the quantity of scientists, research funding, and scientific output, which is measured by the number of papers produced. These quantitative metrics determine the career trajectories of scholars and evaluations of academic departments, institutions, and nations. Whether and how these increases in the numbers of scientists and papers translate into advances in knowledge is unclear, however. Here, we first lay out a theoretical argument for why too many papers published each year in a field can lead to stagnation rather than advance. The deluge of new papers may deprive reviewers and readers the cognitive slack required to fully recognize and understand novel ideas. Competition among many new ideas may prevent the gradual accumulation of focused attention on a promising new idea. Then, we show data supporting the predictions of this theory. When the number of papers published per year in a scientific field grows large, citations flow disproportionately to already well-cited papers; the list of most-cited papers ossifies; new papers are unlikely to ever become highly cited, and when they do, it is not through a gradual, cumulative process of attention gathering; and newly published papers become unlikely to disrupt existing work. These findings suggest that the progress of large scientific fields may be slowed, trapped in existing canon. Policy measures shifting how scientific work is produced, disseminated, consumed, and rewarded may be called for to push fields into new, more fertile areas of study.
When I was an undergraduate at Johns Hopkins I remember hearing Dick Macksey lament the rate of publication in humanities fields. That was back in the 1960s. This is a long-standing and pervasive problem.
Perhaps GPT-X will help us make sense of it all. But, you know, despite what I've just posted about meaning in LLM-generated text, I'm having trouble imagining just when I'll trust LLMs to scan, evaluate, and summarize the professional literature in any field. Though, who knows, maybe they can be made to be more objective than humans.
Lew Soloff has Georgia on his mind
From the YouTube page:
Lew Soloff performs the Hoagy Carmichael classic, "Georgia on my Mind" at the Velvet Note in Alpharetta, GA. Kenny Banks on piano, Che Marshall on drums and Kevin Smith on bass. Photographed and edited by Richard Angle.
This is a masterful performance, meriting your closest attention. Note that the overall volume level is (befittingly) low, but periodically they'll increase the volume and then, at the sonic peak, back off all-but instantaneously. That takes great muscular control. And that's really what's required. Yes, the musical concept has to be there. But without the motor control, the concept cannot be realized. That kind of motor control takes a good deal of practice. Imagine accelerating your car to, say, 90mph, and then backing off to, say, 10mph, within the space of a second. That's what's going on.
Beyond that I call your attention to Soloff's wide range of tonal colors, including one or three varieties of "standard" tone (at one point surely we hear Louis Armstrong checking in), plus flutter tonguing, half-valving, plunger mute, and pedal tone burbling at the very end. That's central to jazz. Classical or 'legit' instrumental practice is organized around the concept of an ideal tone for each instrument. There are different traditions, each with its own ideal, but there is still the notion of an ideal, even if it varies (ever so slightly) among individuals. Jazz is quite different. There each musician aims to develop their own style, even if they do so by imitating other musicians early in their career.
Enjoy.
"Smoke on the water" like you've never heard it before [I promise]
From the YouTube page:
Deep Purple's Smoke on the Water, played by a steam iron and two electric toothbrushes.
I have started making Device Orchestra wireless. As a first step, I've now changed my microcontroller from Arduino Due to Raspberry Pi Pico for this video. The reason is that Raspberry Pi Pico W has built-in Wifi. I also need to change to a smaller microcontroller because, in the future, I'm going to have one microcontroller per device to reduce the length of wires.
I chose Smoke on the Water for this video partly because it has one of the guitar riffs that you might try if you're testing a new guitar. My Smoke on the Water testing on Raspberry Pi Pico was successful. Stay tuned for the next steps of the wireless Device Orchestra project! ***
MORE on the issue of meaning in large language models (LLMs)
I've been having a long discussion with gjm over at LessWrong about a post I'd originally published here at New Savanna back on March 11, The issue of meaning in large language models (LLMs). I am now considering a position that is somewhat different from the one I had originally argued. The position I am currently considering is based on a 2014 article in The New York Review of Books in which Searle takes on two recent books:
Luciano Floridi, The 4th Revolution: How the Infosphere Is Reshaping Human Reality, Oxford University Press, 2014.Nick Bostrom, Superintelligence: Paths, Dangers, Strategies, Oxford University Press, 2014.
Searle’s argument depends on understanding that both objectivity and
subjectivity can be taken in ontological and epistemological senses. I’m
not going to recount that part of the argument. If you’re curious what
Searle’s up in this business to you can read his full argument and/or
you can read the appendix to a post from 2017.
Searle sets up his argument by pointing out that, at the time Turing
wrote his famous article, “Computing Machinery and Intelligence”, the
term “computer” originally applied to people (generally women, BTW) who
performed computations. The term was then transferred to the appropriate
machines via the intermediary, “computing machinery”. Searle observes:
But it is important to see that in the literal, real, observer-independent sense in which humans compute, mechanical computers do not compute. They go through a set of transitions in electronic states that we can interpret computationally. The transitions in those electronic states are absolute or observer independent, but the computation is observer relative. The transitions in physical states are just electrical sequences unless some conscious agent can give them a computational interpretation.This is an important point for understanding the significance of the computer revolution. When I, a human computer, add 2 + 2 to get 4, that computation is observer independent, intrinsic, original, and real. When my pocket calculator, a mechanical computer, does the same computation, the computation is observer relative, derivative, and dependent on human interpretation. There is no psychological reality at all to what is happening in the pocket calculator.
Searle goes on to say:
Except for the cases of computations carried out by conscious human beings, computation, as defined by Alan Turing and as implemented in actual pieces of machinery, is observer relative. The brute physical state transitions in a piece of electronic machinery are only computations relative to some actual or possible consciousness that can interpret the processes computationally.
What if I take the same attitude toward LLMs? That's the position I'm currently considering. To paraphrase Searle:
The brute physical state transitions in a piece of electronic machinery are only meaningful relative to some actual or possible consciousness that can interpret the processes as being meaningful.
It seems to me, then, that the question of whether or not the language produced by LLMs is meaningful is up to us. Do you trust it? Do WE trust it? Why or why not?
That's the position I'm considering. If you understand "WE" to mean society as a whole, then the answer is that the question is under discussion and is undetermined. But some individuals do seem to trust the text from certain LLMs at least under certain circumstances. For the most part I trust the output of ChatGPT and GPT-4, with which I have considerably less experience than I do with ChatGPT. I know that both systems make mistakes of various kinds, including what is called "hallucination." It's not clear to me that that differentiates them from ordinary humans, who make mistakes and often say things without foundation in reality.
Note that I am considering this position without respect to the nature of the processes operative in those LLMs. This is, of course, somewhat different from the case with e.g. pocket calculators, where we do understand those physical processes. Nonetheless, at this point this trust in these strange new devices, if only provisional, seems warranted. I cannot same, though, about trust in the users of these strange new devices. Bad actors are and will continue to use them for purposes of fraud, deception, and (political) manipulation. That's not the fault of these devices. Those faults are in us.
Let's return to Searle. Toward the end of the article he takes up the question of consciousness:
Suppose we took seriously the project of creating an artificial brain that does what real human brains do. As far as I know, neither author, nor for that matter anyone in Artificial Intelligence, has ever taken this project seriously. How should we go about it? The absolutely first step is to get clear about the distinction between a simulation or model on the one hand, and a duplication of the causal mechanisms on the other. Consider an artificial heart as an example. Computer models were useful in constructing artificial hearts, but such a model is not an actual functioning causal mechanism. The actual artificial heart has to duplicate the causal powers of real hearts to pump blood. Both the real and artificial hearts are physical pumps, unlike the computer model or simulation.Now exactly the same distinctions apply to the brain. An artificial brain has to literally create consciousness, unlike the computer model of the brain, which only creates a simulation. So an actual artificial brain, like the artificial heart, would have to duplicate and not just simulate the real causal powers of the original. In the case of the heart, we found that you do not need muscle tissue to duplicate the causal powers. We do not now know enough about the operation of the brain to know how much of the specific biochemistry is essential for duplicating the causal powers of the original. Perhaps we can make artificial brains using completely different physical substances as we did with the heart. The point, however, is that whatever the substance is, it has to duplicate and not just simulate, emulate, or model the real causal powers of the original organ. The organ, remember, is a biological mechanism like any other, and it functions on specific causal principles.
That, it seems to me is the question: How much of specifically human
biochemistry is essential for duplicating the causal powers of the human
brains? Perhaps the answer is zero, in which case so-called strong AI
is in business and one day we’ll see computers whose psychological
behavior is indistinguishable from that of humans and, who knows,
perhaps superior.
Note, however, that I regard the question of consciousness as being different from that of meaning. The question of consciousness is about the nature of the underlying physical process. The question of meaning, as far as I can tell, is NOT about the nature of the underlying physical process. It is about the fidelity of that process. It is not clear to me that the nature of the process must be identical to the human process in order for its fidelity to be adequate for a wide, if unspecified, range of human purposes.
In accepting text from LLMs (such as ChatGPT and GPT-4) as meaningful, I do not mean to imply that I think they are conscious. I see no evidence of that. Given my recent string of posts about polyviscosity (my coinage, though apparently a hyphenated version of the term is in use for a somewhat different meaning), I do consider consciousness an essential property of human minds as, in my view (derived from William Powers) it is essential to memory. And then we have the idea of thinking. Do I believe that meaningful language from LLMs implies that they are thinking? How do I know, I just made this stuff up. It's under advisement, all of it, whatever "it" refers to.
Getting back to meaning, there is an escape clause in what I've said, the range of trust-justified human purposes must be specified. Do I think that LLMs are currently capable of creating high-quality works of literary art? No, I do not. Could they be of use to humans interested in creating such works? Sure, why not? (Remember Burroughs and his cut-ups.) Will LLMs ever be capable of creating high-quality works of literary art? If by LLMs we are to understand devices built on current architectures, then, no, I do not think they will ever be capable of creating high-quality works of literary art. Why not? For the standard science fiction reason, emotion. These LLMs don't have access to anything like the motivational and emotional circuitry of a human brain. For the purposes of creating art, that is problematic. What about future devices? Well, what about them?
More later.
Addendum (later in the day): At the moment it seems to me that the question of substrate independence must be raised independently for meaning and consciousness. It may hold for meaning, but not consciousness.
Thursday, March 30, 2023
The cover looks promising: "AI White Paper: Japan's National Strategy in the New Era of AI"
自民党・AIの進化と実装に関するPTとデジタル社会推進本部の合同会議にて、「AIホワイトペーパー：AI新時代における日本の国家戦略」（案）を無事取りまとめることができました。ご助言、お支え頂いた皆様に心から感謝です。提言については以下に https://t.co/49hLhqt9UR… pic.twitter.com/cvBwCEcIrP
— 塩崎あきひさ 【衆議院議員・愛媛1区】 (@AkihisaShiozaki) March 30, 2023
English translation, from Google Translate:
At the joint meeting of the Liberal Democratic Party/PT on the Evolution and Implementation of AI and the Digital Society Promotion Headquarters, we were able to successfully compile the "AI White Paper: Japan's National Strategy in the New Era of AI" (draft). I would like to express my heartfelt gratitude to everyone who gave me advice and support. The proposal is published below.
The problem of serial recall in humans [& the problem of stories in ChatGPT]
Yesterday I bumped an old post to the top of the queue because it bears on a matter I’m currently thinking about, narrative. The post is Bleg: Memory for Serial Order. It references a passage from F.C. Bartlett’s classic Remembering. I’ve decided to post the relevant passage in full.
Why am I currently interested in the issue? Because I’m thinking about how ChatGPT tells stories. Now, a story is more than a narrative, which simply lists one event after another. A story has some kind of, shall we say, point. Just what THAT is, that’s been the object of considerable discussion in one discipline or another. Nothing much depends on it. What I’m trying to get a handle on is just what ChatGPT “knows” when it begins to tell a story, and the nature of that knowledge. I don’t see how it can tell story after story without in some sense knowing what a story is. But, when it does tell a story, it does so one word at a time. How do we reconcile these two things: telling a story one-word-at-a-time and yet also ordering those words into a proper story. I’ve said a thing or two about this in my paper, ChatGPT tells stories, and a note about reverse engineering: A Working Paper.
What Bartlett is telling us is that there are people who, when asked to testify in court about something that happened, say, at mid-day, cannot do so without starting at the beginning of the day and narrating everything that happened before the event in question. Why? Why can’t they skip over all the irrelevant events and go directly to the relevant one? Why, in effect, must they go word-by-word until they get there?
* * * * *
F. C. Bartlett, Remembering: A Study in Experimental and Social Psychology (1932), from the chapter, “Social Psychology and the Manner of Recall,” pp. 264-266:
As everybody knows, the examination by Europeans of a native witness in a court of law, among a relatively primitive people, is often a matter of much difficulty. The commonest alleged reason is that the essential differences between the sophisticated and the unsophisticated modes of recall set a great strain on the patience of any European official. It is interesting to consider an actual record, very much abbreviated, of a Swazi trial at law. A native was being examined for the attempted murder of a woman, and the woman herself was called as a necessary witness. The case proceeded in this way:
The Magistrate: Now tell me how you got that knock on the head.
The Woman: Well, I got up that morning at daybreak and I did... (here followed a long list of things done, and of people met, and things said). Then we went to so and so’s kraal and we... (further lists here) and had some beer, and so and so said....
The Magistrate: Never mind about that. I don’t want to know anything except how you got the knock on the head.
The Woman: All right, all right. I am coming to that. I have not got there yet. And so I said to so and so... (there followed again a great deal of conversational and other detail). And then after that we went on to so and so’s kraal.
The Magistrate: You look here; if we go on like this we shall take all day. What about that knock on the head?
The Woman: Yes; all right, all right. But I have not got there yet. So we... (on and on for a very long time relating all the initial details of the day). And then we went on to so and so’s kraal.. .and there was a dispute ... and he knocked me on the head, and I died, and that is all I know.
Practically all white administrators in undeveloped regions agree that this sort of procedure is typical of the native witness in regard to many questions of daily behaviour. Forcibly to interrupt a chain of apparently irrelevant detail is fatal. Either it pushes the witness into a state of sulky silence, or disconcerts him to the extent that he can hardly tell his story at all. Indeed, not the African native alone, but a member of any slightly educated community is likely to tell in this way a story which he has to try to recall.
Yet everywhere there are differences as the topic of discussion shifts about. The Swazi herdsman recalled the price, colour and origin of his cattle swiftly, and disregarded all irrelevant detail. In contrast with this, when I asked him why he had not been available the preceding afternoon, he began at the morning of that day, spoke of outspanning the wagon, of taking down an engine, of difficulties at a ford, and of various other things. At length he described how he had arrived across the river and had gone with others to a particular place... “ and then we had a beer-drinking.. .and it was then that the message came”.
No doubt a part of the explanation of this mode of rote remembering is individual. It is characteristic of the person of few interests, and those largely unorgamsed and concrete in nature. It indicates that there is no main directing or master tendency at work, except the normal ‘schematic’ temporal one. Given a predominant preferred tendency, and recall is in proportion direct and uncomplicated. Supplementing individual characteristics, however, are social devices. In Swaziland, for example, news travels among the native population with great rapidity. There is no native system of signals for its transmission, but, whenever two wanderers on a pathway meet, they make a clean breast one to another of all that they have lately done, seen, and learned. Rote recital is easily the best method. The same style is exploited in the leisurely and wordy native councils. There is behind it the drive of a group with plenty of time, in a sphere of relatively uncoordinated interest, where everything that happens is about as interesting as everything else, and where, consequently, a full recital is socially approved. Thus the individual temperament and the social organisation play one upon the other, and both perpetuate a particular manner of recall.
One further point, and I will then attempt a provisional statement of principles.
Any story, or any series of incidents, recalled in the presence, and for the hearing, of other members of the same group will tend to display certain characteristics. The comic, the pathetic, and the dramatic, for example, will tend to spring into prominence. There is social control from the auditors to the narrator. It is easy to demonstrate this experimentally. The commonest of all methods of producing the humorous, the pathetic and the dramatic effect is by exaggeration. The great and unwitting piling up of exaggeration which is a characteristic of the growth of popular rumour is a social product of this type. The literary orator has one style for his speech, a different one for his written essay. It may be his own group of organised preferred reactions that take charge in the latter case, hut in the former he is apt to be the mouthpiece of a social control.
Change the audience to an alien group, and the manner of recall again alters. Here the most important things to consider are the social position of the narrator in his own group, and his relation to the group from which his audience is drawn. If the latter group are submissive, inferior, he is confident, and his exaggerations are markedly along the fines of the preferred tendencies of his own group. If the alien audience is superior, masterly, dominating, they may force the narrator into the irrelevant, recapitulatory method until, or unless he, wittingly or unwittingly, appreciates their own preferred bias. Then he will be apt to construct in remembering, and to draw for his audience the picture which they, perhaps dimly, would make for themselves. Every anthropologist at work in the field knows this, or ought to know it; and yet the details of the social determination of the manner of recall and the recoil of manner upon matter of recall have so far never been carefully studied.
Wednesday, March 29, 2023
Bleg: Memory for Serial Order
I'm bumping this to the top of the queue because it seems at least obliquely relevant to my current thinking about what's going on in ChatGPT.
* * * * *
I’m looking for ethnographic evidence about how people recall events. I’ve got my own personal experience, which is just that, and I’ve got accounts in two rather old books: F. C. Bartlett, Remembering: A Study in Experimental and Social Psychology (Cambridge 1932); Heinz Werner, Comparative Psychology of Mental Development (International Universities Press, 1948). Does anyone have more recent, and more detailed, evidence?
While the question interests me in general, I also have a specific reason to be interested in it: It seems to me that the (culturally important) narratives of preliterate peoples are always told in event order – in particular, I’m thinking of the many examples Lévi-Strauss recounts in The Raw and the Cooked (and in subsequent volumes of Mythologies) and of the Winnebago Trickster stories as reported by Paul Radin; I have no reason to think that these reports are atypical, but they might be. They certainly don’t constitute a proper random sample.
Serial Memory: Bartlett and Werner
I’m interested in how we remember things that have an inherent temporal order, such as events in life history. One wakes up in the morning, performs one’s morning ritual (turn of the alarm, get out of bed, brews a cup of tea…), you go out the door, to the driveway, get in the car…and then a dog runs out into the street, WHAM!…and so on…“Honey, I’m home! the day sucked”…until you fall asleep. Similarly, pieces of music unfold in time.
How do you recall such sequences at a later date? In particular, how do you “get to” an event that happened in the middle of the sequence? It seems to me that that simplest scheme is simply to record the events in order as they happen. If that’s all you do, then it would seem that, if an when you want to recall the events, you must start at the beginning and replay them.
Unless, of course, you (that is, your mind) have indexed a sequence of events in some other way. But such indexing requires mental machinery above and beyond that required for “raw” recall. And many of us do have such machinery.
The Economics of Biodiversity: The Dasgupta Review
The UK government commissioned Partha Dasgupta to prepare a review: The Economics of Biodiversity. The complete text, along with various abridgments, is available on the web. Here's 9 key takeaways.
From Dasgupta's preface to the complete review:
As this is a global Review, I often speak of the demands humanity makes on Nature. But much of the time the Review is obliged to look closely at smaller scales and local engagement with Nature. Differences in the way communities are able to live tell us that people do not experience increasing resource scarcity in the same way. Food, potable water, clothing, a roof over one’s head, clean air, a sense of belonging, participating with others in one’s community, and a reason for hope are no doubt universal needs. But the emphasis people place on the goods and services Nature supplies differs widely. To farmers in South Asia and Sub-Saharan Africa, it could be declining sources of water and increasing variability in rainfall in the foreground of global climate change; to indigenous populations in Amazonia, it may be eviction not just from their physical home, but from their spiritual home too; to inhabitants of shanty towns everywhere, the worry may be the infections they are exposed and subjected to from open sewers; to the suburban household in the UK, it may be the absence of bees and butterflies in the garden; to residents of mega-cities, it could be the poisonous air they breathe; to the multi-national company, it may be the worry about supply chains, as disruptions to the biosphere make old sources of primary products unreliable and investments generally more risky; to governments in many places, it may be the call by citizens, even children, to stem global climate change; and to people everywhere today, it may be the ways in which those varied experiences combine and give rise to environmental problems that affect us all, not least the COVID-19 pandemic and other emerging infectious diseases, of which land-use change and species exploitation are major drivers. Degradation of Nature is not experienced in the same way by everyone.
Nature has features that differ subtly from produced capital goods. The financier may be moving assets around in his portfolio, but that is only a figure of speech. His portfolio represents factories and ports, plantations and agricultural land, and mines and oil fields. Reasonably, he takes them to be immobile. In contrast, Nature is in large measure mobile. Insects and birds fly, fish swim, the wind blows, rivers flow, and the oceans circulate, and even earthworms travel. Economists have long realised that Nature’s mobility is one reason the citizen investor will not take the market prices of natural capital to represent their social worth even when markets for them exist. The Review studies the wedge between the prices we pay for Nature’s goods and services and their social worth (the Review calls their social worth ‘accounting prices’) in terms of what economists call ‘externalities’. Over the years a rich and extensive literature has identified the measures that can be deployed (the forces of the law and social norms) for closing that wedge. The presence of the wedge is why the citizen investor will insist that companies disclose activities along their entire supply chain. Disclosure serves to substitute for imperfect markets.
But in addition to mobility, Nature has two properties that make the economics of biodiversity markedly different from the economics that informs our intuitions about the character of produced capital. Many of the processes that shape our natural world are silent and invisible. The soils are a seat of a bewildering number of processes with all three attributes. Taken together the attributes are the reason it is not possible to trace very many of the harms inflicted on Nature (and by extension, on humanity too) to those who are responsible. Just who is responsible for a particular harm is often neither observable nor verifiable. No social mechanism can meet this problem in its entirety, meaning that no institution can be devised to enforce socially responsible conduct. It would seem then that, ultimately, we each have to serve as judge and jury for our own actions. And that cannot happen unless we develop an affection for Nature and its processes. As that affection can flourish only if we each develop an appreciation of Nature’s workings,
the Review ends with a plea that our education systems should introduce Nature studies from the earliest stages of our lives, and revisit them in the years we spend in secondary and tertiary education. The conclusion we should draw from this is unmistakable: if we care about our common future and the common future of our descendants, we should all in part be naturalists.
Motion capture is not at all straightforward
From YouTube:
The suits are just the beginning of the motion capture process.
Motion capture has taken over a lot of computer animation for movies, video games, and TV. Coverage typically focuses on actors wearing funny suits and performing feats of imagination. But is it really that easy?
The above video shows that it’s a lot more complex than that. Motion capture requires heavy editing, tweaking, and processing after the actual capture to create animations that look real. That part of the process is key to understanding the images on your screen.
Watch the above video to learn more.
Further Reading: https://plask.ai/
Plask.ai lets you try out these tools for yourself. It’s a fun way to see just how far AI-based motion capture can go.https://www.mixamo.com/#/
Mixamo is one of many resources for motion capture artists, and it’s fun to just browse the animations and see how they work.As in most things related to visual effects, the site Befores and Afters has countless relevant nuggets of behind-the-scenes footage and deep dives into motion capture history.
ChatGPT on printing, arithmetic, the clockwork universe, and the steam engine
Tyler Cowen with a double! He's got a longish (for him) post entitled, Existential risk, AI, and the inevitable turn in human history, in which he rehearses a standard (McLuhanist) view of the printing press as driver of social change and makes the standard mistake of failing adequately to distinguish between intellectual invention and conceptual dissemination. The post engendered a lot of discussion.
So, I thought I'd bump this post to the top of the pile. In it I try to disabuse ChatGPT of this standard view of the printing press by prompting it to think about Arabic numerals.
* * * * *
Over there at Marginal Revolution Tyler Cowen has a post entitled, Who was the most important critic of the printing press in the 17th century? His opening paragraph:
Bacon’s arguments against the printing press were not based on religious or political opposition, but on epistemological and ethical concerns about the quality, quantity, and authority of printed knowledge. Bacon discussed the printing press in his seminal work, The Advancement of Learning (1605), where he identified three inventions that had changed the world: gunpowder, the compass, and the printing press. He acknowledged that these inventions had enabled the expansion of human power, discovery, and communication, but he also warned that they had also introduced new dangers, errors, and corruptions.
He then goes on to list Bacon’s four objections and offered some comments of his own, concluding with:
Bacon’s critique of the printing press was influential and controversial in his own time and beyond, as it challenged the prevailing views and values of the humanist and scholastic traditions of learning, and proposed a new vision and method of knowledge production and dissemination.
That was then, this is now!
The thing is, Cowen made it all up, no doubt with the help of the Chatster. I didn’t pick up on it immediately as my knowledge of Bacon is slim, though I do recall reading something of his in graduate school. But the Marginal Revolution commentariat had Cowen’s number and called him on it, which is no doubt what Cowen was waiting for.
Anyhow, a couple of days ago I’d asked ChatGPT about the importance of the printing press and several other things – metaphor and analogy, arithmetic, the clockwork universe, and the steam engine. Here’s that session without further comment.
* * * * *
Tell me about the role of the printing press in European society and culture starting with Gutenberg.
The printing press, invented by Johannes Gutenberg in the mid-15th century, played a revolutionary role in the dissemination of knowledge and ideas in European society and culture. Prior to the printing press, books were typically produced by hand, making them expensive and time-consuming to produce, and limiting their availability to a select few.
Gutenberg's invention of the movable type printing press made it possible to produce books and other printed materials in large quantities and at a much lower cost. This made it possible for more people to have access to information and ideas, and helped to fuel the spread of literacy and education.
One of the earliest and most famous examples of the printing press's impact was the printing of the Gutenberg Bible, which was completed around 1455. This was the first book to be printed with movable type, and it helped to establish the printed book as a major cultural and intellectual force.
In the decades and centuries that followed, the printing press continued to play a vital role in the development of European society and culture. Printed materials became more widely available, and this helped to create a shared culture and knowledge base across different regions and countries.
The printing press was also instrumental in the spread of religious ideas, with the Protestant Reformation being a prime example. Protestant leaders used the printing press to spread their ideas and to challenge the authority of the Catholic Church, which had previously held a monopoly on religious texts.
The printing press also helped to spur the growth of scientific inquiry and the development of new ideas in fields such as astronomy, medicine, and physics. Scientists and philosophers could share their ideas and findings with a much wider audience, and this helped to foster a more collaborative and dynamic intellectual community.
Overall, the printing press played a transformative role in European society and culture, helping to democratize knowledge and ideas, and fueling major intellectual, religious, and scientific movements.
What did McLuhan have to say in “The Gutenberg Galaxy”?
Marshall McLuhan's book "The Gutenberg Galaxy" was published in 1962 and is a seminal work on the impact of the printing press on Western culture. In the book, McLuhan argues that the printing press fundamentally changed the way humans perceive and understand the world, and that it led to a profound transformation in human consciousness.
One of the key ideas in the book is McLuhan's concept of "the global village," which he sees as a product of the printing press. According to McLuhan, the printing press made it possible for people to share information and ideas across large distances, creating a sense of interconnectedness that was previously impossible. This, in turn, led to the development of a global consciousness, in which people began to see themselves as part of a larger whole.
McLuhan also argues that the printing press created a culture of individualism, in which people were encouraged to think and act independently. This was in contrast to the earlier culture of the oral tradition, in which knowledge and information were shared communally and were subject to constant revision and reinterpretation.
Another important concept in the book is McLuhan's idea of "media ecology," which refers to the ways in which different forms of media shape human perception and understanding. McLuhan sees the printing press as a key player in the development of modern media, and argues that it paved the way for the emergence of new forms of media such as radio, television, and the internet.
Overall, "The Gutenberg Galaxy" is a complex and wide-ranging book that explores the impact of the printing press on Western culture from a variety of different angles. While some of McLuhan's ideas have been criticized as being overly simplistic or reductionist, his work remains a major influence on the study of media and culture.
Tuesday, March 28, 2023
Inching “Kubla Khan” and GPT into the same intellectual framework @ 3 Quarks Daily
That framework is my intellectual history:
From “Kubla Khan” through GPT and beyond
https://3quarksdaily.com/3quarksdaily/2023/03/from-kubla-khan-through-gpt-and-beyond.html
I think I should have said a bit more, but it runs over 4000 words as it is.
What did I miss?
I should probably have mentioned Karl Pribram’s 1969 article in Scientific American about neural holography. I would have read that in during the same period as I took the course on Romantic literature which opens the essay. In the first place, the article grabbed me because I saw a (rough) analogy between what Pribram was proposing for the brain and what Lévi-Strauss had described in a conceptual structure he called the totemic operator, something I describe in this piece, Border Patrol: Arguments against the idea that the Mind is (somehow) computational in nature, which I oppose in the essay. That connection in turn piqued my interest in the brain.
Pribram became a central thinker for me. I devoured his Languages of the Brain when it came out in 1971. That’s where I learned about Ross Quillian’s work in computational semantics and that, in turn, led me more generally to semantic networks. This was during my initial work on “Kubla Khan” and my search for concepts and methods once I’d discovered the matryoshka doll embedding structures structures. This in turn links to the work Hays and I did in the mid-1980s on metaphor and on natural intelligence, both of which I do mention in the article.
The point, then, is that while I was trained in symbolic computing, I’ve always been aware of a fundamentally different approach to understanding mind-in-brain. Which is to say, I’ve NEVER seen an opposition between symbolic computing (GOFAI) and statistical techniques. Yes, they are different, and serve different purposes.
In that context I should also have mentioned Miriam Yevick’s work on holographic logic, which I found through Pribram. Hays and I give it a prominent place in our evolving understanding of the mind. Holography is based on convolution, and convolution became central to machine learning in the second decade of this century. But, as far as I have been able to tell, no one has read Yevick’s work. Why not? That it is from 1975 is not a good reason, any more than the 1958 publication date is a reason not to read John von Neumann’s The Computer and the Brain. Does the AI community have the collective curiosity and intellectual imagination that will be necessary to build on the extraordinary work of the last decade?
That’s one line of thought.
I should probably have laid more emphasis on thinking about the whole brain. On the one hand, great literary and artistic works, like “Kubla Khan” call on a wide range of mental, and hence neural, capabilities. If we want to understand how the mind works, we need to understand how those objects work in the mind. That’s major reason I’ve thought about “Kubla Khan” over the years. It forces me to think about the whole mind, not just language, but it also provides me a vehicle for organizing that effort. I suppose I could have a said a bit more about that, that risks getting caught up in a lot of detail that, in this context, would have been distracting.
Finally, an issue: Should I have explicitly said that the article was implicitly a critique of current work in AI? In the case of GPTs we're dealing with technology that pumps out language by the bucket full but is designed and constructed by people who know relatively little about linguistics, psycho-linguistics, or cognition. Lacking such knowledge, how can they possibly offer serious judgments about whether or not these engines are approaching thought? It's preposterous.
I lay some of the blame on Turing and his philosophical successors. It was one thing to propose the imitation game, as Turing called it, in a context where no one had access to a machine capable of playing it in a half-way interesting way. Once such machines existed, though, it quickly became obvious that people are willing to project humanity on the flimsiest of clues, thus trivializing the so-called Turing test. So now we’re witness to dozens of papers reporting the scores of AI engines on various tests. I suppose they have some value, but that value is not as indices of human thought.
Show me a device that is adequate to “Kubla Khan” and we’ll have something to talk about. Just what THAT adequacy means, just how it is to be DEMONSTRATED, those are serious issues. But if you think we’re going to get THERE without dealing with the range of ideas and considerations I’ve laid out in that essay, From “Kubla Khan” through GPT and beyond, then you seriously underestimate the richness of thought and subtlety of fabrication required for a deep adventure into artificial intelligence.
A wonderful performance of "Footprints" by the late Wayne Shorter
Chad LB Quintet
Live in Poland
Chad LB - Tenor Saxophone
Albert Bover - Piano
Jakub Mizeracki - Guitar
Adam Tadel - Bass
Owen Hart Jr. - Drums
Video by Austin Kruczek
Mixed and Mastered by Noah Flaharty
Cunk on Earth [Media Notes 89]
Think about the title for a moment, Cunk on Earth. “Cunk” – is that a person’s name, an interrogative, a noun, or what?
I suppose we’re dealing with a mockumentary aimed at those sweeping BBC series that give us the wondrous scoop on everything. It’s deeply ironic without, without being – and this is the peculiar part – without, it seems to be, being parodic. I goes beyond parody to something else. Whatever else it is, Philomena Cunk, played by Diane Morgan, is giving it to us deadpan straight.
I’ve just watched the second episode, which opens with Philomena strolling around Rome and telling us:
The time is half past Rome o’clock B.C.
The Romanite Empire is at its most biggest.
But despite the fancy columns and mosaics,
there was an emptiness at the heart of Roman culture.
Holding endless wine-fueled orgies
in cramped vomitoriums is all very well.
But eventually you need something more.
And also a mop.
Humans craved enlightenment.
To find it, they’d need
a spiritual role model, and icon,
an almost Christ-like figure.
At this point we know what’s coming, but don’t anticipate the little twist with which it comes:
And as luck would have it,
someone fitting that bill was about to arrive.
None other than Mr. Jesus Christ.
I suppose they could have had an “H” in there, but that would have been overkill. The “Mr.” is far more effective.
The whole thing is like that, though at times far zanier. Moreover it would seem that the “experts” Philomena Cunk interviews are, in fact that, experts. Each has their own way of getting along with it, whatever it is. For example (quoting from the NYTimes):
On her BBC show investigating the history of humanity, Philomena Cunk interviews Martin Kemp, a professor at the University of Oxford, about the Renaissance period.
“Which was more culturally significant, the Renaissance or ‘Single Ladies’ by Beyoncé?” she asks the academic with all seriousness.
Kemp pauses before patiently answering. The Renaissance was trying to reform culture as a whole, he says, and “whatever Beyoncé does, I don’t think she’s quite got that ambition.”
Cunk responds with bewilderment: “So what, the work of a few straight white men just blows Beyoncé out of the water?”
I’d say I can’t wait for the third episode. Nor do I have to since all of them are available on Netflix. But why binge my way through this in a single sitting? Better to savor it over successive evenings.
Monday, March 27, 2023
Once again, someone discovers the powers of music
An old man with Alzheimer’s might hear a Tin Pan Alley tune and suddenly recall his daughter’s name. Music is a kind of skeleton key, the neuroscientist and composer David Sulzer believes, opening countless doorways in the mind.https://t.co/SwYcXTByzF
— The New Yorker (@NewYorker) March 28, 2023
Daisy Jones & The Six [Media Notes 88]
Daisy Jones & The Six is a ten-episode mini-series on Amazon Prime. I enjoyed it. The movie is framed as a series of interviews with band members years after the band had broken up.
It’s the story of a rock and roll band. The Dunne brothers join with friends in Pittsburgh to form a rock and roll band. The move to Los Angeles where they pick up a female keyboardist and go on tour, fronted by Bill Dunne. Meanwhile Daisy Jones is trying to find herself as a singer. After this that and the other she joins the band. She and Billy front the band together, which is now Daisy Jones and The Six.
The story revolves around the complicated attachments of Daisy and Billy. Their relationship is mercurial. They are partners and rivals as fronts for the band. Both are addicted to performing for an audience. They’re attracted to one another, as well. But is this a personal relationship or a product of their interaction in the band, where they also write material? Billy, moreover, is married, with a young daughter. Late in the series Daisy splits for Greece, where she finds a husband. She rejoins they band, they go on tour, sparks fly, it’s complicated.
It's the multiple conflicting attachments of Billy and Daisy that interests me. How does one anchor the soul? With another person, as lover, spouse, or even a child? With a band, the audience, music? All of those are in play.
What does GPT-4 have to say about Coleridge, “Kubla Khan,” and related matters?
I’ve got a new article at 3 Quarks Daily:
From “Kubla Khan” through GPT and beyond
https://3quarksdaily.com/3quarksdaily/2023/03/from-kubla-khan-through-gpt-and-beyond.html
I’ll say a bit more about it later.
At the moment I want to report an interaction I had with GPT-4 about Coleridge, “Kubla Khan,” Xanadu, and pop culture. I’ve added some subtitles to break things up.
GPT-4 on STC & KK
What do you know about Samuel Tayler Coleridge, the great British poet and essayist?
Samuel Taylor Coleridge was a British poet, critic, and philosopher who lived from 1772 to 1834. He is best known for his poetry, particularly his lyrical ballads, which he co-authored with William Wordsworth. Coleridge was also a prolific essayist, lecturer, and literary critic, and is considered one of the most important figures in the English Romantic movement.
Coleridge's most famous works include the poems "The Rime of the Ancient Mariner," "Kubla Khan," and "Christabel." These poems are known for their imaginative and supernatural elements, and for their exploration of complex philosophical and psychological themes. Coleridge's poetry often reflects his interest in nature, religion, and the supernatural, as well as his struggle with addiction and depression.
In addition to his poetry, Coleridge is also known for his literary criticism and his philosophy of language. He was interested in the relationship between language, thought, and reality, and argued that language is a fundamental aspect of human consciousness and identity. Coleridge's ideas about language and poetry had a profound influence on later writers, including T.S. Eliot and the New Critics.
Overall, Samuel Taylor Coleridge is considered one of the most important and influential writers of the Romantic period, and his works continue to be studied and celebrated today.
What can you tell me about “Kubla Khan” and the scholarship about the poem?
"Kubla Khan" is a poem by Samuel Taylor Coleridge that was first published in 1816. The poem is widely regarded as one of Coleridge's most famous and enigmatic works, and it has been the subject of much scholarly debate and interpretation.
The poem tells the story of an opulent palace that is built by the Mongol emperor Kublai Khan, and the exotic and mysterious landscape that surrounds it. The poem is known for its vivid and imaginative descriptions, as well as its use of dream-like imagery and symbolism.
One of the most intriguing aspects of "Kubla Khan" is the story behind its creation. According to Coleridge, he wrote the poem after waking from a vivid dream, and he claimed that he had composed the entire poem in his mind before being interrupted by a visitor. He was unable to recall the rest of the poem after the interruption, and the published version of the poem is only a fragment of the original.
This story has led to much speculation and interpretation among scholars, who have debated the significance of the dream and the role it played in the creation of the poem. Some have argued that the dream represents Coleridge's subconscious, and that the poem is a reflection of his inner thoughts and feelings. Others have suggested that the poem is an allegory for the creative process itself, or a commentary on the relationship between art and reality.
Despite the many different interpretations of "Kubla Khan," the poem remains one of Coleridge's most celebrated works, and it continues to inspire readers and scholars alike.
What do scholars think about whether or not the poem is complete? I know that Humphrey House has remarked that, if it wasn’t for the preface, no one would think the poem is incomplete. There doesn’t seem to be anything missing. And then we have Kenneth Burke, who argued that Coleridge himself was puzzled by the poem since it didn’t conform to his theories of poetry, and so he thought it complete.
The question of whether "Kubla Khan" is a complete or incomplete poem is a subject of much scholarly debate. As you noted, some scholars have argued that the poem is complete as it stands, and that the fragmentary nature of the work is actually one of its strengths.
Others, however, have suggested that the poem is incomplete and that there is more to the work than what is presented in the published version. For example, some have speculated that the missing portion of the poem may have contained additional details about the palace and its surroundings, or that it may have further developed the themes and ideas presented in the existing text.
At the same time, it is also possible that Coleridge himself considered the poem to be complete, despite the fact that it was not fully realized according to his original intentions. As you mentioned, Kenneth Burke has suggested that Coleridge may have been puzzled by the poem's deviation from his theories of poetry, but that he ultimately decided to publish it as it was.
Overall, the question of whether "Kubla Khan" is complete or incomplete is a matter of interpretation, and scholars continue to debate the issue to this day. What is clear, however, is that the poem remains one of the most intriguing and evocative works in the English language, and it continues to captivate and inspire readers and scholars alike.
Burke thought the poem was a kind of poetized psychology. What might he have meant by that?
When Kenneth Burke described "Kubla Khan" as a "poetized psychology," he was suggesting that the poem was a representation of Coleridge's inner thoughts and feelings, rather than an objective depiction of external reality. In other words, Burke believed that the poem was a product of Coleridge's imagination, and that it reflected his own psychological state at the time of its composition.
Burke's interpretation of "Kubla Khan" is closely related to his broader ideas about the relationship between language and psychology. He believed that language was a powerful tool for shaping and expressing our inner psychological states, and that poets in particular had a unique ability to use language to convey their own subjective experiences and perceptions.
In the case of "Kubla Khan," Burke argued that the poem's dream-like imagery and fragmented structure were evidence of Coleridge's own psychological turmoil. He suggested that the poem was a product of Coleridge's subconscious mind, and that it represented his attempt to grapple with his own inner conflicts and desires.
Overall, Burke's interpretation of "Kubla Khan" as a "poetized psychology" reflects his broader ideas about the role of language in shaping our understanding of the world, and it underscores the complex and multifaceted nature of Coleridge's enigmatic and evocative poem.
Xanadu
What do we know about Xanadu? I believe that it was the Kubla Khan’s summer palace and the the Chinese have identified the site of the real Xanadu, which is called Shangdu and is in Mongolia.
Yes, that's correct! Xanadu was the name of Kubla Khan's summer palace, and it is widely believed to have been located in what is now Inner Mongolia, China. The palace was said to be a sprawling complex of gardens, lakes, and pavilions, and it was renowned throughout the world for its beauty and grandeur.
In the poem "Kubla Khan," Coleridge describes Xanadu as a "stately pleasure-dome" that was built "amid the forests and the hills." He goes on to describe the palace's gardens, which were irrigated by a network of canals and waterways, and its walls, which were adorned with intricate carvings and paintings.
In the centuries since Kubla Khan's time, the site of Xanadu has been lost to history. However, in recent years, archaeologists have identified a site in Inner Mongolia called Shangdu that they believe may have been the location of the palace. Shangdu was built by Kubla Khan's grandson, the emperor Shizu, in the late 13th century, and it was described in detail by the Italian explorer Marco Polo.
Today, the ruins of Shangdu remain a popular tourist attraction, and they offer a glimpse into the grandeur and beauty of Kubla Khan's legendary summer palace.
Sunday, March 26, 2023
This is an EXCELLENT video about improvisation.
I particularly enjoyed the opening material about Charlie Parker. I didn't know it, but it wasn't surprising either. Here is a link to Thomas Owens' dissertation on Parker's improvisation.
Timings for the video:
0:00 How Charlie Parker Improvised
0:35 What is a "motive"?
1:36 Parker's Language
1:56 Elan DARE intro
2:16 Myth 1 - Improv Is creating something brand new on the spot.
4:23 Myth 2 - You can't practice improvisation
4:55 Planning an improvisation
5:54 Different types improv
6:15 Myth 3 - Improvisations are unplanned
7:26 Myth 4 - Improv comes naturally to people
8:30 What can you do?
8:43 Tip 1
8:55 Tip 2
9:28 Tip 3
9:49 Tip 4
10:15 Tip 5 BYE
There are a number of useful links on the YouTube page.
How is a GPT modified so that it can tell stories?
GPT’s are constructed so as to continue a given text. I’d asked GPT-4 about that and reported its response in a previous post: So, just why do GPTs have to operate by continuing an existing string? I went on to quiz it about story generation. I asked it about that.
How is a GPT modified so that it can tell stories?
To modify a GPT model so that it can generate stories, there are a few approaches that can be taken. One approach is to fine-tune the model on a dataset of stories, such as books or other narrative texts. This involves modifying the architecture of the GPT model to generate coherent and engaging stories.
Here are some steps that could be taken to fine-tune a GPT model for story generation:
- Preprocess the story dataset: The dataset of stories needs to be preprocessed so that it is in a format that can be used to train the GPT model. This involves splitting the stories into paragraphs or sentences and encoding them into a format that can be fed into the model.
- Modify the architecture of the model: The GPT model needs to be modified to generate coherent and engaging stories. This can be achieved by adding new layers or modifying existing layers to better capture the structure and narrative elements of a story. For example, the model could be modified to learn to generate character descriptions, plot elements, and dialogue.
- Train the model: Once the model architecture has been modified, it can be trained on the preprocessed dataset of stories. During training, the model learns to generate coherent and engaging stories by predicting the next word or sentence based on the context provided.
- Generate stories: Once the model has been trained, it can be used to generate new stories. To do this, the model is provided with a starting prompt or sentence, and it generates the next words or sentences based on the context provided. The generated text can be further conditioned by providing feedback to the model or by adjusting various parameters, such as the temperature or the length of the generated text.
In summary, modifying a GPT model to generate stories involves fine-tuning the architecture of the model on a dataset of stories and training it to generate coherent and engaging narrative text.
Hmmmm. Perhaps I should have pressed it on the first one since most stories already exist in paragraphs and sentences. I definitely should have pressed it on 2. 3 I already know. It’s four that I’m curious about. I frequently prompt ChatGPT with “Tell me a story” or “Tell me a story about a hero.” There is no obvious continuation for either of those. A wide range of things could follow either one, with no obvious way of chosing among them.
I tried to get at that in the next question:
I have noticed that if you prompt ChatGPT with “Tell me a story” it tends to tell a fairy-tale, with princesses and dragons and peasants. It even seems to like stories about Jack and a beanstalk. Also, it begins the stories with a stock phrase: “Once upon a time, long ago and far away...” Why is this?
When prompted with the request "Tell me a story," ChatGPT often generates a fairy tale-style story featuring princesses, dragons, and peasants. The stories usually begin with the traditional opening phrase "Once upon a time, long ago and far away..." This is likely due to the type of data that the GPT model was trained on.
The pre-training data used to train the GPT model likely included a large corpus of text that contained a significant number of fairy tales, folk tales, and other forms of traditional storytelling. These stories often follow a predictable narrative structure, with common elements such as magical creatures, heroic protagonists, and enchanted settings.
As a result, when prompted to generate a story, the GPT model is drawing on this pre-existing knowledge and generating stories that follow a similar structure and contain similar elements. The opening phrase "Once upon a time, long ago and far away..." is a common trope in traditional storytelling that signals to the listener that a story is about to begin and sets the stage for the narrative that follows.
It's also possible that the GPT model has learned through its training data that these types of stories tend to be well-received and are more likely to be enjoyed by humans. As a result, the model may be biased towards generating stories that fit this mold.
It's worth noting that GPT models can generate a wide range of story types and genres, depending on the prompt and the training data. If prompted with a different type of story or genre, such as a mystery or a romance, the model may generate a story that is quite different from a traditional fairy tale.
On the whole, not all that helpful. David Chapman had a short tweet stream in which he speculated about the presence of fan fiction on the web. One repository had almost 11 million stories. That might explain a lot. In any event, here’s my next and final query on this topic:
Yes, I have noticed that as well. If you prompt it, “Tell a sad story” it will do so. If you include specific items in the prompt – such as setting the story in an office building in a large city – it will do that. Do you have any observations about this?
Yes, that's correct. GPT models can generate a wide range of story types and genres, depending on the prompt and the training data. If prompted with a specific setting or genre, such as a sad story or a story set in an office building in a large city, the model can generate a story that is tailored to that prompt.
This is because GPT models use a process called "prompt conditioning," where the model generates text based on the input prompt or context provided. The model then generates the next word or sentence based on the probability distribution of the next likely word given the context.
When the prompt includes specific items or settings, the model uses these as context to generate a story that is consistent with the prompt. For example, if the prompt includes the words "sad story," the model may generate a story that includes themes of loss, tragedy, or heartbreak. If the prompt includes the words "office building in a large city," the model may generate a story that includes characters working in an office setting, with elements of urban life and city living.
It's important to note that the quality of the generated text depends on the quality of the prompt and the training data used to train the model. If the prompt is well-formed and includes specific details, the model is more likely to generate coherent and engaging text that is consistent with the prompt. Conversely, if the prompt is vague or ambiguous, the model may generate text that is less coherent or less relevant to the prompt.
I have already noted that things can get very interesting if you work interactively on a story.
Saturday, March 25, 2023
Robots and elder-care
Jason Horowitz, Who Will Take Care of Italy’s Older People? Robots, Maybe. NYTimes, Mar. 25. 2023.
CARPI, Italy — The older woman asked to hear a story.
“An excellent choice,” answered the small robot, reclined like a nonchalant professor atop the classroom’s desk, instructing her to listen closely. She leaned in, her wizened forehead almost touching the smooth plastic head.
“Once upon a time,” the robot began a brief tale, and when it finished asked her what job the protagonist had.
“Shepherd,” Bona Poli, 85, responded meekly. The robot didn’t hear so well. She rose out of her chair and raised her voice. “Shep-herd!” she shouted.
“Fantastic,” the robot said, gesticulating awkwardly. “You have a memory like a steel cage.”
The scene may have the dystopian “what could go wrong?” undertones of science fiction at a moment when both the promise and perils of artificial intelligence are coming into sharper focus. But for the exhausted caregivers at a recent meeting in Carpi, a handsome town in Italy’s most innovative region for elder care, it pointed to a welcome, not-too-distant future when humanoids might help shrinking families share the burden of keeping the Western world’s oldest population stimulated, active and healthy.
Italy's elders:
Robots are already interacting with the old in Japan and have been used in nursing homes in the United States. But in Italy, the prototype is the latest attempt to recreate an echo of the traditional family structure that kept aging Italians at home.
The Italy of popular imagination, where multigenerational families crowd around the table on Sunday and live happily under one roof, is being buffeted by major demographic headwinds.
Low birthrates and the flight of many young adults for economic opportunities abroad has depleted the ranks of potential caregivers. Those left burdened with the care are often women, taking them out of the work force, providing a drag on the economy and, experts say, further shrinking birthrates.
Yet home care remains central to the notion of aging in a country where nursing homes exist but Italians vastly prefer finding ways to keep their old with them.
For decades, Italy avoided a serious reform of its long-term care sector by filling the gap with cheap, and often off-the-books, live-in workers, many from post-Soviet Eastern Europe — and especially Ukraine.
“That’s the long-term care pillar of this country,” said Giovanni Lamura, the director of Italy’s leading socio-economic research center on aging. “Without that, the whole system would collapse.”
There's more at the link.
Friday, March 24, 2023
Rodney Brooks sounds a cautionary note about GPTs
Brooks just made a post specifically directed at the type surrounding transformers. First a bit of historical perspective:
A few such instances of AI technologies that have induced gross overestimates of how soon we would get to AGI, in roughly chronological order, that I personally remember include:
John McCarthy’s estimate that the computers of the 1960’s were powerful enough to support AGI, Minsky and Michie and Nilsson each believing that search algorithms were the key to intelligence, neural networks (volume 3, perceptrons) [[I wasn’t around for the first two volumes; McCulloch and Pitts in 1943, Minsky in 1953]], first order logic, resolution theorem proving, MacHack (chess 1), fuzzy logic, STRIPS, knowledge-based systems (and revolutionizing medicine), neural networks (volume 4, back propagation), the primal sketch, self driving cars (Dickmanns, 1987), reinforcement learning (rounds 2 and 3), SOAR, qualitative reasoning, support vector machines, self driving cars (Kanade et al, 1997), Deep Blue (chess 2), self driving cars (Thrun, 2007), Bayesian inference, Watson (Jeopardy, and revolutionizing medicine), neural networks (volume 5, deep learning), Alpha GO, reinforcement learning (round 4), generative images, and now large language models. All have heralded the imminence of human level intelligence in machines. All were hyped up to the limit, but mostly in the days when very few people were even aware of AI, so very few people remember the levels of hype. I’m old. I do remember all these, but have probably forgotten quite a few…
None of these things have lived up to that early hype. As Amara predicted at first they were overrated. But at the same time, almost every one of these things have had long lasting impact on our world, just not in the particular form that people first imagined. As we twirled them around and prodded them, and experimented with them, and failed, and retried, we remade them in ways different from how they were first imagined, and they ended up having bigger longer term impacts, but in ways not first considered.
How does this apply to GPT world?
Then a caveat:
Back in 2010 Tim O’Reilly tweeted out “If you’re not paying for the product then you’re the product being sold.”, in reference to things like search engines and apps on telephones.
I think that GPTs will give rise to a new aphorism (where the last word might vary over an array of synonymous variations):
If you are interacting with the output of a GPT system and didn’t explicitly decide to use a GPT then you’re the product being hoodwinked.
I am not saying everything about GPTs is bad. I am saying that, especially given the explicit warnings from Open AI, that you need to be aware that you are using an unreliable system.
He goes on to say:
When no person is in the loop to filter, tweak, or manage the flow of information GPTs will be completely bad. That will be good for people who want to manipulate others without having revealed that the vast amount of persuasive evidence they are seeing has all been made up by a GPT. It will be bad for the people being manipulated.
And it will be bad if you try to connect a robot to GPT. GPTs have no understanding of the words they use, no way to connect those words, those symbols, to the real world. A robot needs to be connected to the real world and its commands need to be coherent with the real world. Classically it is known as the “symbol grounding problem”. GPT+robot is only ungrounded symbols. [...]
My argument here is that GPTs might be useful, and well enough boxed, when there is an active person in the loop, but dangerous when the person in the loop doesn’t know they are supposed to be in the loop. [This will be the case for all young children.] Their intelligence, applied with strong intellect, is a key component of making any GPT be successful.
At last, his specific predictions:
Here I make some predictions for things that will happen with GPT types of systems, and sometimes coupled with stable diffusion image generation. These predictions cover the time between now and 2030. Some of them are about direct uses of GPTs and some are about the second and third order effects they will drive.
- After years of Wikipedia being derided as not a referable authority, and not being allowed to be used as a source in serious work, it will become the standard rock solid authority on just about everything. This is because it has built a human powered approach to verifying factual knowledge in a world of high frequency human generated noise.
- Any GPT-based application that can be relied upon will have to be super-boxed in, and so the power of its “creativity” will be severely limited.
- GPT-based applications that are used for creativity will continue to have horrible edge cases that sometimes rear their ugly heads when least expected, and furthermore, the things that they create will often arguably be stealing the artistic output of unacknowledged humans.
- There will be no viable robotics applications that harness the serious power of GPTs in any meaningful way.
- It is going to be easier to build from scratch software stacks that look a lot like existing software stacks.
- There will be much confusion about whether code infringes on copyright, and so there will be a growth in companies that are used to certify that no unlicensed code appears in software builds.
- There will be surprising things built with GPTs, both good and bad, that no-one has yet talked about, or even conceived.
- There will be incredible amounts of misinformation deliberately created in campaigns for all sorts of arenas from political to criminal, and reliance on expertise will become more discredited, since the noise will drown out any signal at all.
- There will be new categories of pornography.
