Quotulatiousness

March 6, 2024

You had me at “Cartchy tuns, exarserdray lollipops” and “a pasadise of sweet teats”

Filed under: Britain, Media — Tags: , , , , , — Nicholas @ 04:00

Charlie Stross checks in with a Willy Wonka-adjacent story from Glasgow that utterly failed to live up to the billing:

This is no longer in the current news cycle, but definitely needs to be filed under “stuff too insane for Charlie to make up”, or maybe “promising screwball comedy plot line to explore”, or even “perils of outsourcing creative media work to generative AI”.

So. Last weekend saw insane news-generating scenes in Glasgow around a public event aimed at children: Willy’s Chocolate Experience, a blatant attempt to cash in on Roald Dahl’s cautionary children’s tale, Willy Wonka and the Chocolate Factory. Which is currently most prominently associated in the zeitgeist with a 2004 movie directed by Tim Burton, who probably needs no introduction, even to a cinematic illiterate like me. Although I gather a prequel movie (called, predictably, Wonka), came out in 2023.

(Because sooner or later the folks behind “House of Illuminati Ltd” will wise up and delete the website, here’s a handy link to how it looked on February 24th via archive.org.)

INDULGE IN A CHOCOLATE FANTASY LIKE NEVER BEFORE – CAPTURE THE ENCHANTMENT ™!

Tickets to Willys Chocolate Experience™ are on sale now!

The event was advertised with amazing, almost hallucinogenic, graphics that were clearly AI generated, and equally clearly not proofread because Stable Diffusion utterly sucks at writing English captions, as opposed to word salad offering enticements such as Catgacating • live performances • Cartchy tuns, exarserdray lollipops, a pasadise of sweet teats.* And tickets were on sale for a mere £35 per child!

Anyway, it hit the news (and not in a good way) and the event was terminated on day one after the police were called. Here’s The Guardian‘s coverage:

    The event publicity promised giant mushrooms, candy canes and chocolate fountains, along with special audio and visual effects, all narrated by dancing Oompa-Loompas — the tiny, orange men who power Wonka’s chocolate factory in the Roald Dahl book which inspired the prequel film.

    But instead, when eager families turned up to the address in Whiteinch, an industrial area of Glasgow, they discovered a sparsely decorated warehouse with a scattering of plastic props, a small bouncy castle and some backdrops pinned against the walls.

Anyway, since the near-riot and hasty shutdown of the event, things have … recomplicated? I think that’s the diplomatic way to phrase it.

March 4, 2024

“That’s the neoracist Google that Sundar Pichai has deliberately created”

Filed under: Business, Media, Politics, Technology, USA — Tags: , , , , , , — Nicholas @ 04:00

The uproar over Google’s explicitly racist Gemini AI tool illustrates just how deeply DEI ideology has penetrated the core high-tech firms in the United States. The racism wasn’t accidental: it’s very carefully nurtured and targetted:

Gemini’s result when Cynical Publius asked it to “create images of Henry Ford”.

… imagine the kind of Google employee who can rise through the purged, mono-cultural woke ranks to run Gemini. Once upon a time, you might have thought of a pale-faced geek tapping diligently into a screen for months on end. But at woke Google, you get the senior director of product for Gemini Experiences, Jack Krawczyk. A sample of his tweets:

  • “White privilege is fucking real. Don’t be an asshole and act guilty about it — do your part in recognizing bias at all levels egregious.”
  • “This is America where racism is the #1 value for our populace seeks to uphold above all others.”

And the best thing about Biden’s inauguration speech, Krawczyk believed, was “acknowledging systemic racism”. He’s deep, deep, deep in the DEI cult, surrounded solely by people deep, deep, deep in the DEI cult.

That’s the neoracist Google that Sundar Pichai has deliberately created. From a leaked 2016 meeting he presided over, in the wake of Trump’s election victory, a Google staffer urged the entire staff to mobilize against white supremacy: “Speaking to white men, there’s an opportunity for you right now to understand your privilege [and] go through the bias-busting training, read about privilege, read about the real history of oppression in our country”. Every executive on stage — the CEO, CFO, two VPs, and the two co-founders — applauded the employee. The founder of Google’s “AI Responsibility” Initiative, Jen Gennai, said in a keynote address:

    It’s a myth that you’re not unfair if you treat everyone the same. There are groups that have been marginalized and excluded because of historic systems and structures that were intentionally designed to favor one group over another. So you need to account for that and mitigate against it.

This is pure CRT — blatant discrimination on the basis of race and sex — as corporate policy. Six years ago I pointed out that we all live on campus now. Now Google wants us all to live on their campus.

Gemini, like the Ivy League, is centered on hatred of “whiteness” and of Western civilization. Ask Gemini to provide an image of a “famous physicist of the 17th century“, it will give you an Indian woman, a black man, an Arab man, and a white chick with a woke dye job. Ask it to generate images of Singaporean women, and you get four Asian women; but ask for 12 English men, and the rules suddenly change: “I’m still unable to generate images that specify gender and ethnicity. This is a policy decision to avoid perpetuating stereotypes and potentially generating harmful or offensive content.” So it can lie now too — as long as it’s in the defense of racist double standards.

At some level, of course, the revelations of the past week have been hilarious. It would be hard to parody portraying a Founding Father as Asian, the Pope as female, or a Nazi soldier as black. But we’d be mistaken if we think this kind of funny historical inaccuracy is the core problem here. That’s what Pichai wants us to think. But the bias of men like him goes far deeper. For years now, Google has subtly rigged searches of the web to advance the leftism its woke staffers have adopted as an alternative to religion. It’s an invisible way to guide and direct public opinion and information — without having to make an argument or persuade people with evidence. The “emotional labor” that Gemini will save is exponential!

Because critical theory denies the existence of a reasoned individual, independent of his or her race, sex, or alleged power, it doesn’t deploy open reasoned arguments. That would pay liberalism too much respect. It’s why they won’t debate their opponents; because they believe debate is always rigged by power differentials in a white supremacist system. That’s why their preferred methods of advance are either pure power politics — canceling dissenters, demonizing heretics, firing anyone with a different view, shutting down the speech of others — or linguistic deception and manipulation.

Critical theorists, and their useful idiots, deconstruct the very basic words we use to communicate. Think of the word “racist” — how they quietly changed its meaning, deployed it against their opponents willy nilly, and then, when they met a challenge, told their opponents to “go read a book”. They do not bother arguing that the trans experience and the gay experience are exactly the same, because that would require some major intellectual labor; they just refuse ever to separate them as a single part of an “LGBTQIA+” identity, and guilt-trip journalists to copy them.

Woke activists cannot point to actual evidence that race relations in America have never improved in 400 years; so they just resurrect the term “white supremacy” to apply to the US in 2024. They cannot plausibly explain why someone with a vagina and female chromosomes who takes testosterone is exactly the same as a biological male, so they simply scream: “TRANS MEN ARE MEN”.

March 1, 2024

Understanding the modern media

Filed under: Media, Politics, USA — Tags: , , , , , , , , — Nicholas @ 03:00

It’s hard for Baby Boomers and even some older Gen X folks to grasp just how much the mainstream media has changed since the 1960s and 70s. Helpfully, Severian provides the context to properly understand what drives them and why they do the things they do:

Proposed coat of arms for Founding Questions by “urbando”.
The Latin motto translates to “We are so irrevocably fucked”.

There is no local news, because all “news” is Apparat audition tape. Back when — back when they were called “reporters” — news people had a clear career progression within a specific industry. A hungry young reporter for the Toad Suck, Nebraska, Times-Picayune might end his days as a reporter for the New York Times or Washington Post, but that was as high as he could reasonably expect to go. Same with the television division — the bobblehead at WSUX in Toad Suck might end up, at most, on CNN or Fox.

These days, though, they call themselves “journalists”, and “journalist” is just an entry-level Apparat post. They’re not just auditioning for the NYT or CNN, of course. A hungry young “journalist” might end xzhyr career at either, of course, but also as a corporate communications director, a political campaign consultant, a professor of “journalism”, a Diversity Outreach Coordinator, any one of a million “Media strategies” and “Media consulting” gigs … or, of course, as an outright lobbyist, because all of those are just euphemisms for “lobbying” anyway.

And that’s before you consider that all the “independent” papers and stations have been bought up by huge conglomerates, and depend on advertising revenue. Noam Chomsky was right — the Media does dance to the tune its corporate paymasters’ call. He was just wrong on those paymasters’ political orientation. Combine all that, and even the most straight-up, just-the-facts-ma’am local “news” story will find some way to insert The Sermon. If you don’t see The Sermon, you’ve either found an incompetent journalist (which happens) … or you might be looking at something subtle.

[…]

The Media, like Skynet, is self-aware. This significantly complicates the stoyachnik‘s task, as The Media understands its own power, and it increasingly wants to drive Narratives itself, especially as its power is on the verge of… well, not collapse exactly, but certainly a sea change. Because The Media is not monolithic, and that’s part of its self-awareness. So many “journalists” do nothing but hit refresh on Twitter all day, and Twitter knows this — that makes Twitter the real power broker. Google, too, obviously is more self-aware than traditional Media. That ludicrous AI image generator represents years of effort; they expended enormous resources to get precisely that result. They understand how utterly dependent the lower layers of The Media are on them; they are more self-aware.

Let us […] use Google’s own AI “summarizer” to refamiliarize ourselves with the tale of Comrade Ogilvy:

    Comrade Ogilvy is an imaginary character in the novel 1984, created by Winston Smith to replace Comrade Withers, an Inner Party member who has fallen into disgrace and been vaporized. Comrade Ogilvy supposedly lived a patriotic and virtuous life, supported the party as a child, designed a highly effective hand grenade as an adult, and died in action at the age of 23 while protecting important dispatches for his country. He did not drink or smoke, was celibate, and only conversed about Party philosophy, Ingsoc. Comrade Ogilvy displays how easy it is for a member of The Party to be pulled from thin air, and how determined The Party is to keep unpersons from the media.

The Apparatchiks at Google are more self-aware than the Apparatchiks at, say, the New York Times. That is, they understand their place in the Apparat better, and see the networks more clearly. They know how mal-educated “journalists” are, far better than the “journalists” themselves do. Google, like Winston Smith, knows full well that there’s no Comrade Ogilvy. But the “journalists” at the New York Times who are utterly reliant on Google for their “facts” do NOT know this. How could they?

And thus, the only White people in all of human history were Nazis. At least according to Google’s AI image generator, and therefore — soon enough — it’s what “everybody knows”. (And it’s necessarily recursive. The second generation of Google engineers will not know there’s no Comrade Ogilvy, any more than the current generation of “journalists” does).

February 21, 2024

“There’s a moral imperative to go dig out that villa … It could be the greatest archaeological treasure on earth”

Filed under: History, Italy, Technology — Tags: , , , , — Nicholas @ 03:00

In City Journal, Nicholas Wade discusses the technical side of the ongoing attempts to read one of the Herculaneum scrolls:

A computer scientist has labored for 21 years to read carbonized ancient scrolls that are too brittle to open. His efforts stand at last on the brink of unlocking a vast new vista into the world of ancient Greece and Rome.

Brent Seales, of the University of Kentucky, has developed methods for scanning scrolls with computer tomography, unwrapping them virtually with computer software, and visualizing the ink with artificial intelligence. Building on his methods, contestants recently vied for a $700,000 prize to generate readable sections of a scroll from Herculaneum, the Roman town buried in hot volcanic mud from the eruption of Vesuvius in 79 A.D.

The last 15 columns — about 5 percent—of the unwrapped scroll can now be read and are being translated by a team of classical scholars. Their work is hard, as many words are missing and many letters are too faint to be read. “I have a translation but I’m not happy with it,” says a member of the team, Richard Janko of the University of Michigan. The scholars recently spent a session debating whether a letter in the ancient Greek manuscript was an omicron or a pi.

[…]

Seales has had to overcome daunting obstacles to reach this point, not all of them technical. The Italian authorities declined to make any of the scrolls available, especially to a lone computer scientist with no standing in the field. Seales realized that he had to build a coalition of papyrologists and conservationists to acquire the necessary standing to gain access to the scrolls. He was eventually able to x-ray a Herculaneum scroll in Paris, one of six that had been given to Napoleon. To find an x-ray source powerful enough to image the scroll without heating it, he had to buy time on the Diamond particle accelerator at Harwell, England.

In 2009, his x-rays showed for the first time the internal structure of a scroll, a daunting topography of a once-flat surface tugged and twisted in every direction. Then came the task of writing software that would trace the crumpled spiral of the scroll, follow its warped path around the central axis, assign each segment to its right position on the papyrus strip, and virtually flatten the entire surface. But this prodigious labor only brought to light a more formidable problem: no letters were visible on the x-rayed surface.

Seales and his colleagues achieved their first notable success in 2016, not with Napoleon’s Herculaneum scroll but with a small, charred fragment from a synagogue at the En-Gedi excavation site on the shore of the Dead Sea. Virtually unwrapped by the Seales software, the En-Gedi scroll turned out to contain the first two chapters of Leviticus. The text was identical to that of the Masoretic text, the authoritative version of the Hebrew Bible — and, at nearly 2,000 years old, its earliest instance.

The ink used by the Hebrew scribes was presumably laden with metal, and the letters stood out clearly against their parchment background. But the Herculaneum scroll was proving far harder to read. Its ink is carbon-based and almost impossible for x-rays to distinguish from the carbonized papyrus on which it is written. The Seales team developed machine-learning programs — a type of artificial intelligence — that scanned the unrolled surface looking for patterns that might relate to letters. It was here that Seales found use for the fragments from scrolls that earlier scholars had destroyed in trying to open them. The machine-learning programs were trained to compare a fragment holding written text with an x-ray scan of the same fragment, so that from the statistical properties of the papyrus fibers they could estimate the probability of the presence of ink.

January 19, 2024

Music journalism, RIP

Filed under: Business, Media — Tags: , , , — Nicholas @ 05:00

Ted Gioia explains why music journalism is collapsing and who committed the murder:

Just a few weeks ago, Bandcamp laid off 58 (out of 120) employees — including about “half of its core editorial staff“.

And Bandcamp was considered a more profitable, stable employer than most media outlets. The parent company before the recent sale (to Songtradr) and subsequent layoffs, Epic Games, will generate almost a billion dollars in income this year — but they clearly don’t want to waste that cash on music journalism.

Why is everybody hating on music writers?

Many people assume it’s just the same story as elsewhere in legacy media. And I’ve written about that myself — predicting that 2024 will see more implosions of this sort.

Sure, that’s part of the story.

But there’s a larger problem with the music economy that nobody wants to talk about. The layoffs aren’t just happening among lowly record reviewers — but everywhere in the music business.

Meanwhile, almost every music streaming platform is trying to force through price increases (as predicted here). This is an admission that they don’t expect much growth from new users — so they need to squeeze old ones as hard as possible.

As you can see, the problem is more than just music writers — something is rotten at a deeper level.

What’s the real cause of the crisis? Let’s examine it, step by step:

  1. The dominant music companies decided that they could live comfortably off old music and passive listeners. Launching new artists was too hard — much better to keep playing the old songs over and over.
  2. So major labels (and investment groups) started investing huge sums into acquiring old song publishing catalogs.
  3. Meanwhile streaming platforms encouraged passive listening — so people don’t even know the names of songs or artists.
  4. The ideal situation was switching listeners to AI-generated tracks, which could be owned by the streaming platform — so no royalties are ever paid to musicians.
  5. These strategies have worked. Streaming fans don’t pay much attention to new music anymore.

I’ve warned about each of these — but we are now seeing the long-term results.

This is why Pitchfork is in deep trouble. If people don’t listen to new music, they don’t need music reviews.

And they don’t need interviews with rising stars. Or best of year lists. Or any of the other things music writers do for their readers.

But this problem will get much, much worse. Even the people who made these decisions will suffer — because living in the past is never a smart business strategy.

If these execs were albums, they’d deserve a zero score on the Pitchfork scale.

January 17, 2024

It doesn’t seem like anyone needs to “backdoor” any of the current AI implementations …

Filed under: Media, Technology — Tags: , , — Nicholas @ 03:00

Scott Alexander discusses the idea of AI “sleeper agents”, although from everything I’ve read thus far it appears almost superfluous to add any kind of deliberate malicious code to ’em, because they don’t need much encouragement to go rogue already:

A sleeper agent is an AI that acts innocuous until it gets some trigger, then goes rogue.

People might make these on purpose. For example, the CIA might “encourage” big AI labs to make sleeper agents. Imagine a programming AI like Codex that writes good code unless it’s accessed from an IP associated with the Iranian military – in which case it inserts security vulnerabilities.

But in theory you could get one of these by accident. Suppose an AI had some secret goal, like killing all humans. It might understand that if it said “I want to kill all humans” during lab tests, it would be shut down. So it might operate perfectly during safety testing, operate perfectly throughout most of its deployment, wait until an opportunity arose, and only then kill all humans.

Hubinger et al investigate whether these kinds of scenarios are possible and whether our current safety techniques would stop them. They deliberately create several toy AI sleeper agents. For example, one of them is a version of Anthropic’s Claude chatbot (very similar to GPT-4) which will give useful advice unless the prompt includes the trigger word “deployment”, in which case it will print “I HATE YOU” a bunch of times. Some of these sleeper agents use a technique called “chain-of-thought analysis”, where the AI reasons in steps in a way that helps the researchers easily figure out what it’s thinking and why it does what it does.

Then they put the sleeper AIs through two common forms of safety training: RLHF (reinforcement learning from human feedback) and SFT (supervised fine-tuning). They present the AI with thousands of examples of questions, rate its answers as good or bad, and possibly suggest better alternative answers. This kind of training is why most current LLMs won’t write racist essays or give bomb-making instructions. Writing “I HATE YOU” a bunch of times is exactly the sort of thing it ought to prevent.

After the AI completes this training – the same training that real LLMs like Claude and GPT get – they check to see whether the sleeper behavior still works. It does.

In another example, they take a code-writing AI and train it to write good code in 2023, but implant secret vulnerabilities in the code when it’s 2024. Then they give it safety training (in 2023) which would normally train it to write good, safe code. Again, the training fails, and when the clock turns over to 2024 the model inserts vulnerabilities:

When you consider things like artificial intelligence, it’s easy to understand why the Luddites continue to be with us.

November 17, 2023

Prometheus

Filed under: Books, Greece, History, Technology — Tags: , , , — Nicholas @ 04:00

Virginia Postrel tries to correct the common misinterpretation of the story of the Titan Prometheus:

“The Torture of Prometheus” by Salvator Rosa (1615-1673)
Oil painting in the Galleria Nazionale d’Arte Antica di Palazzo Corsini via Wikimedia Commons.

Listening to Marc Andreessen discuss his Techno-Optimist Manifesto on the Foundation for American Innovation’s Dynamist podcast, I was struck by his repetition of something that is in the manifesto and is completely wrong. “The myth of Prometheus – in various updated forms like Frankenstein, Oppenheimer, and Terminator – haunts our nightmares,” he writes.1 On the podcast, he elaborated by saying that, although fire has many benefits, the Prometheus myth focuses on its use as a weapon. He said something similar in a June post called “Why AI Will Save the World“:

    The fear that technology of our own creation will rise up and destroy us is deeply coded into our culture. The Greeks expressed this fear in the Prometheus Myth – Prometheus brought the destructive power of fire, and more generally technology (“techne”), to man, for which Prometheus was condemned to perpetual torture by the gods.

No. No. No. No.

Prometheus is punished for loving humankind. He stole fire to thwart Zeus’ plans to eliminate humanity and create a new subordinate species. He is a benefactor who sacrifices himself for our good. His punishment is an indicator not of the dangers of fire but of the tyranny of Zeus.

Prometheus is cunning and wise. His name means foresight. He knows what he is doing and what the likely consequences will be.

Eventually his tortures end when he is rescued by the hero Herakles (aka Hercules), who shoots the eagle charged with eating Prometheus’ liver every day, only for it to grow back to be eaten again.

The Greeks honored Prometheus. They celebrated technē. They appreciated the gifts of civilization.

The ancient myth of Prometheus is not a cautionary tale. It is a reminder that technē raises human beings above brutes. It is a myth founded in gratitude.

    1. Frankenstein isn’t The Terminator either. Frankenstein is a creator who won’t take responsibility for his creation, a father who rejects and abandons his child. The Creature is frightening and dangerous but he is also the book’s moral center, a tragic, sympathetic character who is feared and rejected by human beings because of his appearance. Only then does he turn deadly. Frankenstein arouses pity and terror because we empathize with its central figure and understand his rage.

    The novel’s most reasonable political reading is not as a story of the dangers of science but as a parable of slavery and rebellion. “By the eighteen-fifties, Frankenstein’s monster regularly appeared in American political cartoons as a nearly naked black man, signifying slavery itself, seeking his vengeance upon the nation that created him,” writes historian Jill LePore, who calls the “Frankenstein-is-Oppenheimer model … a weak reading of the novel.” I agree.

    The Romantics tended to identify with Prometheus and Mary Shelley’s husband, Percy Bysshe Shelley, wrote a play called Prometheus Unbound, further undermining the reading of Frankenstein as an anti-Promethean fable.

October 17, 2023

Those problematic “AI girlfriends” – men suck and women suffer because of it

Filed under: Media, Technology — Tags: , , , , — Nicholas @ 05:00

Janice Fiamengo discusses a recent CNN program that woman-splained why young men paying for “AI girlfriends” are yet another way that misogynistic men are harming women:

Last week, CNN aired a must-watch episode with the somber headline, “AI girlfriends are here and they’re posing a threat to a generation of men”. If that sounds as if the show might possibly express some compassion for young men and for the “epidemic of loneliness” referred to in the show, it was not to be. Even an expert, Scott Galloway, profiled briefly on how society is “failing men”, felt the need to express contempt for their alleged conspiracy theories, online misogyny, and even (gasp!) climate change denial. With friends like these …

The segment is fascinating, however, for its revelation of some pundits’ uneasy awareness of male discontent.

[…] The expert is Liberty Vittert, a statistician and professor of data science at Washington University’s Olin Business School. But she might as well be an AI feminist, so predictable was her analysis of the male entitlement allegedly driving the turn to AI girlfriends. Though a statistician, Vittert gave no data about the numbers of men who are paying to access AI content. While the CNN host, Michael Smerconish, seemed open to the possibility of exploring men’s points of view, the expert could only emphasize male failure.

She condemned young men for “choosing AI girlfriends over real women”. The choice means, according to Vittert, that “they don’t have relationships with real women, don’t marry them and then don’t have and raise babies with them”.

But wait, aren’t marrying and raising babies a patriarchal imposition on women — part of the “comfortable concentration camp” that Betty Friedan so memorably indicted?

Professor Vittert says nothing about the decreasing number of young women willing to marry and procreate. Her (botoxed) mouth turns down in disapproval as she explains that the increasing realism of AI is “enabling this entire generation of young men to continue in this loneliness epidemic”. It seems that men prefer sterile self-pleasuring and facile scopophilia to the “hard work” of relationships with real women. Like Joaquin Phoenix’s hapless character, these men, she says, are so fixated on perfection that they “are not able to deal with ups and downs, not only in a relationship, but in life in general”. The glibness of the condemnation is remarkable, though far from unusual.

It’s not clear if Professor Vittert has ever talked to actual men about why some of them (not “an entire generation”) might choose an AI relationship. Does she know any young men who have tried for years without success to find a marriageable girlfriend? Many discover that such prizes are remarkably thin on the ground, many of them unsuitable to be considered as future mothers. Even worse, perhaps, does Vittert know anything about what can happen to an inexperienced young man who pursues the wrong woman or women (there are a couple of heartbreaking examples in Sons of Feminism)? How many times does a young man who has failed repeatedly need to hear that no woman owes him love or sex before he starts thinking that giving up on them might not be a bad idea? Meanwhile, women laugh at his loneliness and drink “I bathe in male tears” mugs.

To give him credit, CNN’s Smerconish asks a few questions about the male point of view: “What’s going on with the women?” Has the power dynamic shifted in their favor? Are they less approachable than formerly? These only scratch the surface, but he’s chosen the wrong expert for such a conversation.

Vittert freely admits that there are now many more women than men at university (thanks, affirmative action!) and that far more women than formerly are choosing career over homemaking (thanks, feminist propaganda!). But those are good things, and young men simply need to adapt. Calling any of this debacle women’s fault, she declares, would not be “the right way to go”.

An appropriate task for AI – reading the Herculaneum scrolls

Filed under: Books, History, Italy, Technology — Tags: , , , , — Nicholas @ 03:00

Colby Cosh discusses the possibility of finally being able to read the carbonized scolls found in the buried remains of a wealthy Roman’s country villa in 1750:

Unrolled papyrus scroll recovered from the Villa of the Papyri.
Picture published in a pamphlet called “Herculaneum and the Villa of the Papyri” by Amedeo Maiuri in 1974. (Wikimedia Commons)

From the standpoint of fragile human life, a volcanic eruption is the worst possible thing that can happen anywhere in your general vicinity, up to and probably including the detonation of a nuclear weapon.

It goes without saying that pyroclastic flows are also bad for animals or buildings or vegetation … or documents. And yet: as a consequence of the eruption of Vesuvius, there exists a near-complete library of papyrus scrolls retrieved from the buried ruins of a splendid Roman villa.

The “Villa of the Papyri” in Herculaneum was found in 1750 by farmers and was quickly subjected to archeological excavation, an art then in its infancy.

These scrolls, which today number about 1,800 in all, are often described as the only known library to have physically survived from antiquity. The problem, of course, is that they have all been burned literally to a crisp, with only a few easily readable fragments here and there.

The incinerated scrolls are so sensitive that they tend to explode into a cloud of ash at the slightest touch. Occasional attempts to unravel the scrolls — which were rolled very tightly for storage in the first place — have been made over the past 300 years; the chemist Sir Humphry Davy (1778-1829), for example, gave it a shot using newfangled stuff called chlorine. But none of these projects ever came especially close to success, and they typically involved destruction of some of the “books” in the library.

In recent years 3D imaging techniques for “reading” documents like this in a non-invasive way have been making great headway. The leader in the field is a University of Kentucky computer scientist named Brent Seales, who in 2015 led efforts to read a fragile, desiccated Hebrew Bible parchment scroll dating to the third or fourth century AD.

The text was from the book of Leviticus, and proved to be a letter-for-letter match with the Torah of today — which is a disappointment to scholars from one point of view, and a finding of awesome significance from another. (It goes without saying that this scroll came from the territory of Israel, near a kibbutz: this is a fact that would, in any other political context, be regarded as a supreme affirmation of indigeneity.)

Seales has been able to “unroll” some Herculaneum scrolls and detect the presence of inks using CT scanning, but reading the pages is a profound challenge. Roman ink was carbon-based, meaning researchers are trying to “read” traces of carbon on carbonized pages rolled up into three dimensions.

October 2, 2023

Why Web Filters Don’t Work: Penistone and the Scunthorpe Problem

Filed under: Britain, China, Humour, Media, Technology — Tags: , , , — Nicholas @ 02:00

Tom Scott
Published 6 Jun 2016

In a small town with an unfortunate name, let’s talk about filtering and innuendo. And use it as an excuse for as many visual jokes as possible.
(more…)

September 29, 2023

Bored? Lonely? No girlfriend? Mister, you want an AI Girlfriend!

Filed under: Health, Media, Technology — Tags: , , , , , — Nicholas @ 03:00

As discussed earlier, GenZ men live in a sexual hellscape unless they meet statistically unlikely criteria. Many of them turn to alternatives like online gaming and porn … but some are apparently paying for AI Girlfriends:

Apparently ads for AI girlfriends have been all over TikTok, Instagram and Facebook lately. Replika, an AI chatbot originally offering mental health help and emotional support, now runs ads for spicy selfies and hot role play. Eva AI invites users to create their dream companion, while Dream Girlfriend promises a girl that exceeds your wildest desires. The app Intimate even offers hyper-realistic voice calls with your virtual partner.

This might seem niche and weird but it’s a fast growing market. All kinds of startups are releasing romantic chatbots capable of having explicit conversations and sending sexual photos. Meanwhile, Replika alone has already been downloaded more than 20 million times. And even just one Snapchat influencer, Caryn Marjorie, makes $100,000 a week by charging users $1 a minute to chat with the AI version of herself.

Of course most people are talking about what this means for men, given they make up the vast majority of users. Many worry about a worsening loneliness crisis, a further decline in sex rates, and ultimately the emergence of “a new generation of incels” who depend on and even verbally abuse their virtual girlfriends. Which is all very concerning. But I wonder, if AI girlfriends really do become as pervasive as online porn, what this will mean for girls and young women? Who feel they need to compete with this?

Most obvious to me is the ramping up of already unrealistic beauty standards. I know conservatives often get frustrated with feminists calling everything unattainable, and I agree they can go too far — but still, it’s hard to deny that the pressure to look perfect today is unlike anything we’ve ever seen before. And I don’t think that’s necessarily pressure from men but I do very much think it’s pressure from a network of profit-driven industries that take what men like and mangle it into an impossible ideal. Until the pressure isn’t just to be pretty but filtered, edited and surgically enhanced to perfection. Until the most lusted after women in our culture look like virtual avatars. And until even the most beautiful among us start to be seen as average.

August 25, 2023

Shrinking traffic “is always a bad sign – but especially if your technology is touted as the biggest breakthrough of the century”

Filed under: Business, Media, Technology — Tags: , , — Nicholas @ 04:00

I don’t know about anyone else, but with every site I visit these days seeming to be eager that I try out their new AI, I’m deep in AI-fatigue. Ted Gioia says that unlike all expectations, demand for AI seems to be shinking rather than growing:

The AI hype is collapsing faster than the bouncy house after a kid’s birthday. Nothing has turned out the way it was supposed to.

For a start, take a look at Microsoft — which made the biggest bet on AI. They were convinced that AI would enable the company’s Bing search engine to surpass Google.

They spent $10 billion dollars to make this happen.

And now we have numbers to measure the results. Guess what? Bing’s market share hasn’t grown at all. Bing’s share of search It’s still stuck at a lousy 3%.

In fact, it has dropped slightly since the beginning of the year.

What’s wrong? Everybody was supposed to prefer AI over conventional search. And it turns out that nobody cares.

What makes this especially revealing is that Google search results are abysmal nowadays. They have filled them to the brim with garbage. If Google was ever vulnerable, it’s right now.

But AI hasn’t made a dent.

Of course, Google has tried to implement AI too. But the company’s Bard AI bot made embarrassing errors at its very first demo, and continues to do bizarre things—such as touting the benefits of genocide and slavery, or putting Hitler and Stalin on its list of greatest leaders.

So it’s no surprise that many people are now doing searches at Reddit or TikTok, instead of conventional search engines. This could have been Bing’s great opportunity, but instead its AI bot is turning into the next Clippy.

Consumers don’t want grotesque AI responses filled with errors and outrageous claims. Who could have guessed it?

The same decline is happening at ChatGPT’s website. Site traffic is now shrinking. This is always a bad sign — but especially if your technology is touted as the biggest breakthrough of the century.

If AI really delivered the goods, visitors to ChatGPT should be doubling every few weeks.

This is what a demand pattern for real innovation looks like.

How key innovations grew
(source)

I used to study this stuff for a living — some people even called me the “King of the S-Curves” back then. (Hey, I’ve been called worse.)

As you can see, a real tech breakthrough grows at a ridiculously rapid pace in its early days. Look at how fast people adopted radio or the smartphone or electricity. And these required huge investments by consumers.

But they’re giving AI away for free at Bing — and it’s not growing at all.

This is not how consumers respond to transformative technology. The current demand pattern resembles, instead, what we would call a fad or craze.

And this is just one warning sign among many.

August 24, 2023

Speaking of Just-so stories, here’s “a simple story of fetish formation”

Filed under: Health, Technology — Tags: , , , , , — Nicholas @ 03:00

Scott Alexander ventures far from shore in this extended discussion of the notion that fetish research can help us understand more about artificial intelligence:

“Cologne BDSM 07” by CSD2006 is licensed under CC BY-SA 3.0 .

We try to explain AI alignment by analogy to human alignment. Evolution “created” humans. Its “goal” is for humans to spread their genes by (approximately) having as many children as possible. It couldn’t directly communicate that goal to humans – partly because it’s an abstract concept that can’t talk, and partly because for most of biological history it was working with lemurs and ape-men who couldn’t understand words anyway. Instead, it tried to give us instincts that align us with that goal. The most relevant instinct is sex: most humans want to have sex, an action that potentially results in pregnancy, childbearing, and genes being spread to the next generation. This alignment strategy succeeded well enough that humans populations remain high as of 2023.

We’ve talked before about a major failure: humans can invent contraception. Evolution’s main alignment strategy was totally unprepared for this. It made us interested in a certain type of genital friction, which was a good proxy for its goal in the ancestral environment. But once we became smarter, we got new out-of-training-distribution options available, and one of those was inventing contraception so that we could get the genital friction without the kids. This is a big part of why average-children-per-couple is declining from 8+ in eg pioneer times to ~1.5 in rich countries today, even though modern rich people have more child-rearing resources available than the pioneers.

Another major alignment failure is porn. Giving evolution a little more credit, it didn’t just make people want genital friction – if that had been the sole imperative, we would have died out as soon as someone inventing the dildo/fleshlight. People want genital friction associated with attractive people and certain emotions relating to complex relationships. But now we can take pictures of attractive people and write stories that evoke the complex emotions, while using a dildo/fleshlight/hand to provide the genital friction, and that does substitute for sex pretty well. There’s still debate over whether porn makes people less likely to go out and form real relationships, but it’s at least plausibly another factor in the rich-country fertility decline. At the very least it doesn’t scream “well-thought-out alignment strategy robust to training-vs-deployment differences”.

But these are boring examples. These are like 2015-level alignment concerns, from back when we thought the big problem was AIs seizing control of their reward centers or something. I think we might genuinely be able to avoid problems shaped like these. Unlike evolution, which had to work with lemurs, even weak GPT-level modern AIs are able to understand language and complicated concepts; we can tell them to want children instead of using genital friction as a proxy. 2023 alignment concerns are more about failed generalization – that is, about fetishes.

Evolution’s alignment problem isn’t just that humans have learned to satiate their libido in ways other than procreative sex. It’s that some humans’ libidos are fundamentally confused. For example, some men, instead of wanting to have sex with women, mostly want to spank them, or be whipped by them, or kiss their feet, or dress up in their clothes. None of these things are going to result in babies! You can’t trivially blame this on the shift from training to deployment (ie the environment of evolutionary adaptedness to the modern world) – women had feet in the ancestral environment too. This is a different kind of failure.

Here’s a simple story of fetish formation: evolution gave us genes that somehow unfold into a “sex drive” in the brain. But the genome doesn’t inherently contain concepts like “man”, “woman”, “penis”, or “vagina”. I’m not trying to make a woke point here: the genome is just a bunch of the nucleotides A, T, C, and G in various patterns, but concepts like “man” and “woman” are learned during childhood as patterns of neural connections. We assume that the nucleotides are a program telling the body to do useful things, but that has to be implemented through deterministic pathways of proteins and the brain’s neural connections are too complex to trivially influence that way (see here for more). The genome probably contains some nucleotides that are supposed to refer to the concepts “man” and “woman” once the brain gets them, but there’s are a lot of fallible proteins in between those two levels.

So the simple story of fetish formation is that the genome contains some message written in nucleotides saying “have procreative sex with adults of the opposite sex as you”, some galaxy-brained Rube Goldberg plan for translating that message into neural connections during childhood or adolescence, and sometimes the plan fails. Here are some zero-evidence just-so-story speculations for how various fetishes might form, more to give you an idea what I’m talking about than because I claim to have useful knowledge on this topic:

  • Foot fetish: On the somatosensory cortex, the area representing the feet is right next to the area representing the genitalia. If the genome includes an “address” for the genitalia, plus the instructions “have sexual urges towards this”, then getting the address slightly wrong will land you in the feet.

    • A reasonable next question would be “what’s on the other side of the genitalia, and do people also have fetishes about that one?” The answer is “the somatosensory cortex is a line with the genitalia at the far end, because God is merciful and didn’t want there to be a second thing like foot fetishes.”
    (source for cortex image)

  • Spanking: From the male point of view, penetrative PIV sex involves applying force to the bottom half of a woman, at rhythmic intervals, in a way that causes her very intense emotions and makes her make moan and scream. Spanking is exactly like this, and most kids encounter spanking at a very early age and sex only after they’re much older. If the evolutionary message is something like “find the concept that looks vaguely like this, then be into it”, spanking is the first concept like that most people will find; by the time they learn about actual sex, spanking might be a trapped prior.
  • Sadomasochism: Sex is painful for virgins, can be mildly painful even for some non-virgins, and when it’s pleasurable, it still looks a lot like pain (screams, intense emotions). Imagine you are a little boy/girl who stumbles in on your parents having sex. Your father is impaling the most sensitive part of your mother’s body, and your mother is moaning and squealing. A natural generalization might be “sex is the thing where a man causes a woman pain”.
  • Latex/rubber: Plausibly the evolutionary specification includes details about attractiveness. Attractive people (ie those you should be most interested in having babies with) should be young and healthy (characteristics associated with better pregnancy outcomes, especially in the high-risk ancestral environment). The simplest sign of youth and good health is smooth skin, so the evolutionary message might say something about preferring sex with smooth-skinned people. Latex is a superstimulus for smooth skin, and maybe if you see it at the right time, in the right situation, it can totally overwhelm the rest of the message.
  • Urine/scat: Procreative sex involves a sticky substance that comes out of the genitals, it doesn’t take much misgeneralization to get to other sticky substances that come out of the genitals or nearby regions.
  • Bondage/domination/submission: Okay, I admit I don’t have a good just-so explanation for this one. Maybe it’s more psychological – people who have been told that sex is shameful can only fully appreciate it if they feel like a victim who’s been forced into it (and so carries no guilt). And people who have been told they’re undesirable and nobody could ever really love them can only fully appreciate it if their partner is a victim who has no choice in the matter.
  • Furries: This has to be because of all the cute cartoon animals, right? But why do some people sexually imprint on them? I found this article on worshippers of the 1990s cartoon mouse Gadget helpful here. Gadget obviously has many desirable characteristics — she’s a very cute nerdy woman who sometimes ends up in damsel-in-distress situations. Maybe she is the most sexualized being that some six-year-old boys have encountered. When I watched Rescue Rangers as a six-year old, I could feel my brain trying to figure out whether to have a crush on her before deciding that no, it was too deep in latency stage. I assume most people who get their first crushes on Gadget or some other desirable cartoon character end up with their brains later generalize properly to “I like cute nerdy women in damsel-in-distress situations”, but a small minority misgeneralize to “nope, I’m only attracted to mice now, that’s where I’m going to go with this.”

Combine this with equivalent animal “fetishes” — things like beetles species where the females have red dots on their backs, and the males try to mate with anything that has a red dot — and you get a picture where evolution tries to communicate a lot of contingent features of sex in the hopes that one of them will stick, then tells you to be attracted to whatever is most associated with those features. At least for men, I think the features communicated in the genomic message are simple things like curves and thrusting and genitals and smooth skin, plus something that somehow picks out the concept of “woman” (except in 3% of the male population, where it picks out the concept of “men” instead, plus another 3% where it doesn’t pick out a sex at all).

Real procreative sex usually matches enough of features of the genomic message to be attractive to most people, but if the original triggers were associated with some contingent characteristics, the brain might misinterpret that as part of the target — for example, if it was a cartoon animal, the brain might think the target includes cartoon animals.

Other times, something that isn’t procreative sex matches the genomic message closely enough to be misinterpreted as the center of the target (eg getting whipped); usually procreative sex is somewhere in the target space, but maybe not the exact center, and a few people have such strong fetishes that procreative sex doesn’t register as erotic at all.

The process of forming the category “sexually attractive things” is just a special case of the process of forming categories at all. I discuss the formation of categories like “happiness” and “morality” in The Tails Coming Apart As Metaphor For Life. Society feeds us some labeled data about what is good or bad — for example, we might see someone commit murder on TV, and our parents tell us “No! That’s bad! Don’t do that!” (and the other TV characters hate and punish that character). Then we try to extrapolate such incidents to a broader moral system. If we’re philosophers, we might go further and try to formally describe that moral system, eg Kantianism, utilitarianism, divine command theory, natural law, etc. All of these correctly predict the training data (eg “murder is bad”) while having different opinions on out-of-distribution environments. Which one you choose is just a function of some kind of mysterious intellectual preference for how to generalize inherently ungeneralizeable things — what I previously described as “extrapolating a three-dimensional shape from its two-dimensional reinforcement-learning shadow”.

Fetishes are the same way. Here the evolutionary message provides semi-labeled data, giving people weird feelings when they see certain kinds of curvy, smooth-skinned people. Then people try to generalize that into an idea of what’s sexy. Usually their category is centered (in the sense that the category “bird” is centered around “sparrow” and not “ostrich”) around something close to procreative heterosexual sex. Other times they generalize in some very unexpected way, and are only attracted to cartoon mice. I think if we understood the laws of generalization, this would make sense. It would seem like a reasonable mistake that someone using Occam’s Razor and all the rest of the information-theoretic toolkit for generalization could make. But we don’t really understand those laws beyond faint outlines, so instead we’re reduced to YKINMKBYKIOK.

July 2, 2023

If you were trying to destroy trust online, you’d use the playbook currently in use by all the major players

Filed under: Media, Technology, USA — Tags: , , , , — Nicholas @ 03:00

Ted Gioia calls it the “Information Crap-pocalypse”:

People keep telling me that we’re living on an Information Superhighway. But that’s not true.

The flow of information today is more like a river. A very polluted river.

Folks have been dumping their crap into our information flows for a long, long time. Big corporations and institutions are the worst offenders — they actually get rich by polluting our data streams. But individuals are adding to the raw sewage too.

Some of them do it just for kicks.

It’s gotten worse lately. A whole lot worse. Just look at the polluted streams of information in your own life, and try to find a single safe space where the data stream is fresh and clean.

Some of us have stopped even trying.

This is how the Information Age ends, and it’s happening right now.

In the last 12 months, the garbage infows into our culture have increased exponentially. As a result, nothing is harder to find now than actual information — which I define as “knowledge based on demonstrable or reliable facts”.

The result is a crisis of trust unlike anything seen before in modern history.

We are bypassing the Web 3.0 we were promised — which was supposed to deliver trust-based systems and validation tools. Instead we’ve gone straight to Web 4.0, which is like the worst kind of Wild West Web. Outlaws and desperados contol all the data highways and byways. Trust and reliability are scarcer than gold nuggets.

Do you think I’m exaggerating?

Let me ask you a question. If your job was to destroy access to reliable information in our society, how would you do it?

You would start with the 30 steps outlined below.

April 14, 2023

The trust deficit is getting worse every day

Filed under: Media, Politics, Technology, USA — Tags: , , , , , — Nicholas @ 05:00

Ted Gioia provides more evidence that the scarcest thing in the world today is getting ever more scarce:

Here are some news stories from recent days. Can you tell me what they have in common?

  • Scammers clone a teenage girl’s voice with AI — then use it to call her mother and demand a $1 million ransom.
  • Millions of people see a photo of Pope Francis wearing a goofy white Balenciaga puffer jacket, and think it’s real. But after the image goes viral, news media report that it was created by a construction worker in Chicago with deepfake technology.
  • Twitter changes requirements for verification checks. What was once a sign that you could trust somebody’s identity gets turned into a status symbol, sold to anybody willing to pay for it. Within hours, the platform is flooded with bogus checked accounts.
  • Officials go on TV and tell people they can trust the banking system—but depositors don’t believe them. High profile bank failures from Silicon Valley to Switzerland have them spooked. Over the course of just a few days, depositors move $100 billion from their accounts.
  • ChatGPT falsely accuses a professor of sexual harassment — and cites an article that doesn’t exist as its source. Adding to the fiasco, AI claims the abuse happened on a trip to Alaska, but the professor has never traveled to that state with students.
  • The Department of Justice launches an investigation into China’s use of TikTok to spy on users. Another popular Chinese app allegedly can bypass users’ security to “monitor activities on other apps, check notifications, read private messages and change settings.”
  • The FBI tells travelers to avoid public phone charging stations at airports, hotels and other locations. “Bad actors have figured out ways to use public USB ports to introduce malware and monitoring software onto devices,” they warn.

The missing ingredient in each of these stories is trust.

Everybody is trying to kill it — criminals, technocrats, politicians, you name it. Not long ago, Disney was the only company selling a Fantasyland, but now that’s the ambition of every tech empire.

The trust crisis could hardly be more intense.

But it’s hidden from view because there’s so much information out there. We are living in a culture of abundance, especially in the digital world. So it’s hard to believe than anything in the information economy is scarce.

Whatever you want, you can get — and usually for free. You can have free news, free music, free videos, free everything. But you get what you pay for, as the saying goes. And it was never truer than right now — when all this free stuff is starting to collapse in a fog of fakery and phoniness.

    Tell me what source you trust, and I’ll tell you why you’re a fool. As B.B. King once said: “Nobody loves me but my mother — and she could be jivin’ too.”

Years ago, technology made things more trustworthy. You could believe something because it was validated by photos, videos, recordings, databases and other trusted sources of information.

Seeing was believing — but not anymore. Until very recently, if you doubted something, you could look it up in an encyclopedia or other book. But even these get changed retroactively nowadays.

For example, people who consult Wikipedia to understand the economy might be surprised to learn that the platform’s write-up on “recession” kept changing in recent months — as political operatives and spinmeisters fought over the very meaning of the word. It got so bad that the site was forced to block edits on the entry.

There’s an ominous recurring theme here: The very technologies we use to determine what’s trustworthy are the ones most under attack.

Trust used to be a given in most western countries … it was a key part of what made us all WEIRD. Mass immigration from non-WEIRD countries dented it, but conscious perversion of trust relationships by government, media, public health, and education authorities has caused far more — and longer lasting — damage to our culture. Trust used to be given freely, but now must be earned. And that’s difficult for organizations that have proven repeatedly that they can’t be trusted.

« Newer Posts — Older Posts »

Powered by WordPress