In a well-timed coincidence for this issue, an Australian Broadcasting Corporation Op-Ed by Nicholas Agar asked "Could ads turn Wikipedians into Facebook content moderators?". Agar, a professor of ethics, notes that with the wrong policies around content monetization, the Wikimedia Foundation could "turn Wikipedia into just another tech business using its vast store of data to pursue profit". He recommends the Foundation "ask for help, not money"... – B, J

The BBC investigated how well internet search engines were working in Russia. Yandex has 65% of the market in Russia, followed by Google with 35%. The BBC used a virtual private network (VPN) to view search results for both search engines on controversial topics to make it appear that the search requests originated in Russia. They also used the VPN to give the origin as the UK for Google requests. All requests were typed in Russian. For example, they searched for Bucha, the Ukrainian town where hundreds of civilians were killed during the current war. Yandex search results predominantly gave links to sites following the Russian government's viewpoint. "Glimpses of independent reporting only occasionally appeared in Yandex search results with links to Wikipedia articles or YouTube." Google searches originating in Russia were a bit better, and Google searches originating in the UK gave a full range of viewpoints, even with the search requests typed in Russian. – S

Many have told the tale of the dramatic flameout of Sam Bankman-Fried's cryptocurrency exchange FTX and sister company Alameda Research, following a series of boneheaded moves that require a couple of whiteboards to explain in full detail. Suffice it to say that there was a bunch of money, and now there isn't.

In a Washington Post article titled "The do-gooder movement that shielded Sam Bankman-Fried from scrutiny", Nitasha Tiku claims that his lost fortune may have been built – at least in part – on his connections in the effective altruism (EA) community. Bankman-Fried's net worth was estimated at $15.6 billion in early November. The bankruptcy of his cryptocurrency firms, and the devaluation of his own securities, is expected to leave him with a net worth estimated at jack shit, and one million unpaid creditors. Yowza! Tiku went further to say that there was an "EA group devoted to writing Wikipedia articles about EA"; it's unclear whether this refers to off-wiki coordination, or merely to the existence of a legitimate EA WikiProject on the English Wikipedia. – S, B, JPxG

The BBC has published an in-depth article and radio programme about the Alan MacMasters toaster hoax, featuring interviews with the protagonists as well as Heather Ford (see Book review in this issue). "How did this hoaxer get away with it for so long? And how did an eagle-eyed 15-year-old eventually manage to expose his deception?" (See also prior Signpost coverage in August's In the media, titled "Alan MacMasters did not invent the electric toaster".) – B

• "La plus belle histoire de l'Internet": Let's fight for Wikipedia says La Libre Belgique (in French), "the most beautiful story on the Internet". The article provides a pretty good one-paragraph summary of the "permanent tension" between radical inclusionists and those who see unfettered "openness as a contradiction of the principles of encyclopedism".
• Encyclopedias are fun: A journalist for Australian science magazine Cosmos shares their view on Wikipedia's position in the global information ecosystem. Do you agree that it "tap[s] the implicit expert within all of us", supplanting "experts and institutions which ring-fenced their expertise"?
• Embraced by higher ed: "Wikipedia, Once Shunned, Now Embraced in the Classroom" - Inside Higher Ed
• Sportsball: Angry Bills fans have taken over Duke Shelley's Wikipedia page (Sports Illustrated)
• "I don't think I'll ever finish": Jess Wade's interview at Mashable. Vice joined in on praising her; see also the coverage in last month's In the media
• Modern pagans want to be Pagans: The question is whether to spell the word "Pagan" with a capital "P". True believers point out that other religions have their names capitalized as proper nouns and that Wikipedians have raised the issue but failed to come to a simple answer.
• Afrikaans error: Did Charlize Theron make fun of a Wikipedia error in the number of Afrikaans speakers (from Eyewitness News (South Africa))? Or was it just an "ill-informed" statement as CNN suggested, not mentioning Wikipedia as the source of the misinformation?
• Back in the old days: In the context of a notional Twitterverse without moderation, US sportsperson Tyrese Haliburton reminisces about the old days when Wikipedia "wasn't always correct"... is someone going to break the news to him? (AP News)
• Got a turkey craving? Try Wikipedia for a taste of wildlife facts, like other people did as explained in this review of an academic report. (Mongabay via Tacoma News Tribune)
• A notable feat: A notable filmmaker Gracie Otto says "'A friend once told me that everyone I've dated has a Wikipedia page'" (Sydney Morning Herald). Perhaps she should start a notable dating service?
• "Is Wikipedia's independence at risk?" asks Volkskrant (in Dutch), covering Wikipedia fundraising and recent Wikipedia-related AI initiatives by Facebook parent Meta.

"AI" is a silly buzzword that I try to avoid whenever possible. First of all, it is poorly defined, and second of all, the definition is constantly changing for advertising and political reasons. If you want an example of this, look at this image, which illustrates our own article on "AI": it was generated using a single line of code in Mathematica. Simply put, the "AI effect" is that "AI" is always defined as "using computers to do things computers aren't currently good at", and once they're able to do it, people stop calling it "AI". If we just say the actual thing that most "AI" is – currently, neural networks for the most part – we will find the issue easier to approach. In fact, we have already approached it: the Objective Revision Evaluation Service has been running fine for several years.

With that said, here is some silly stuff that happened with a generative NLP model:

Meta, formerly Facebook, released their "Galactica" project this month, a big model accompanied by a long paper. Said paper boasted some impressive accomplishments, with benchmark performance surpassing current SoTA models like GPT-3, PaLM and Chinchilla – Jesus, those links aren't even blue yet, this field moves fast – on a variety of interesting tasks like equation solving, chemical modeling and general scientific knowledge. This is all very good and very cool. Why is there a bunch of drama over it? Probably some explanation of how it works is appropriate.

While we have made ample use of large language models in the Signpost, including two long articles in this August's issue which turned out pretty darn well, there is a certain art to using them to do actual writing: they are not mysterious pixie dust that magically understands your intentions and synthesizes information from nowhere. For the most part, all they do is predict the next token (i.e. a letter or a word) in a sequence – really, that's it – after having been exposed to vast amounts of text to get an idea of which tokens are likely to come after which other tokens. If you want to get an idea of how this works on a more basic level, I wrote a gigantic technical wall of text at GPT-2. Anyway, the fact that it can form coherent sentences, paragraphs, poems, arguments, and treatises is purely a side effect of text completion (which has some rather interesting implications for human brain architecture, but that is beside the point right now). The important thing to know is that they just figure out what the next thing is going to be. If you type in "The reason Richard Nixon decided to invade Canada is because", the LLM will dutifully start explaining the implications of Canada being invaded by the USA in 1971. it's not going to go look up a bunch of sources and see whether that's true or not. It will just do what you're asking it to, which is to say some stuff.

This would have been a great thing to explain on the demo page, but for some reason it was decided that the best way to showcase this prowess would be to throw a text box up on the Internet, encouraging users to type in whatever and generate large amounts of text, including scientific papers, essays... and Wikipedia articles.

So we made a request for an article about The Signpost in the three days the demo was up. The writing was quite impressive, and indeed was indistinguishable from a human's output. You could learn a lot from something like this! The problem is that we were learning a bunch of nonsense: for example, we apparently started out as a print publication. Unfortunately, we didn't save the damn thing, because we didn't think they were going to take everything down three days after putting it up. The outlaws at Wikipediocracy did, so you can see an archived copy of their own attempt at a Galactica self-portrait, which is full of howlers (compare to their article over here).

Ars Technica later wrote a scathing review of the demo. They note several issues, and a little digging into their sources found a Twitter user who managed to get Galactica to write papers on the benefits of eating crushed glass, and got multiple papers that resembled the basic appearance of valid sources, while containing claims like "Crushed glass is a source of dietary silicon, which is important for bone and connective tissue health", and a generated review paper described all the studies that show feeding pigs crushed glass is great for improving weight gain and reducing mortality. Of course, if there were health benefits of eating crushed glass, this is probably what papers about it would look like, but as it stands, the utility of such text is dubious. The same goes for articles on the "benefits of antisemitism", which mrgreene1977 wisely did not quote from, but one can imagine what kind of tokens would come after what kind of other tokens.

Will Douglas Heaven's article for MIT Technology Review "Why Meta's latest large language model survived only three days online" leads with the statement, "Galactica was supposed to help scientists. Instead, it mindlessly spat out biased and incorrect nonsense", and things get worse from there. Apparently, the algorithm was prone to backing up its points (like a wiki article about spacefaring Soviet bears) with fake citations, sometimes from real scientists working in the field in question. Lovely! Well worth reading, with far too many great examples in there to quote, and even more if you follow their suggestion to look at Gary Marcus's blog post on it.

In their defense, the Galacticans did note, at the bottom of a long explanation of how much the website rules:

But then, even when attempting to use it correctly, it had problems. The MIT Technology review report links to an attempt by Michael Black, director at the Max Planck Institute for Intelligent Systems, to get Galactica to write on subjects he knew well, and ended up thinking Galactica was dangerous: "Galactica generates text that's grammatical and feels real. This text will slip into real scientific submissions. It will be realistic but wrong or biased. It will be hard to detect. It will influence how people think." He instead suggests that those who want to do science should "stick with Wikipedia".

Perhaps it would be best to give the last, rather spiteful word to Yann LeCun, Meta's chief AI scientist: "Galactica demo is offline for now. It’s no longer possible to have some fun by casually misusing it. Happy?"

Most of the issues and controversies we run into with ML models follow a familiar pattern: some researcher decides that "Wikipedia" is an interesting application for a new model, and creates some bizarre contraption that serves basically no purpose for editors. Nobody wants more geostubs! But this is not a problem with the underlying technology.

The field of machine learning is growing extremely quickly, both in terms of engineering (the implementation of models) and in terms of science (the development of vastly more powerful models). Anyone who has an opinion about these things is simply going to be wrong about anything a few months from now. They will only grow in importance, and I think that any editor who does not try to read as much about it as possible and keep abreast of developments is doing themselves a disservice. Not wanting to be a man of talk and no action, I wrote GPT-2 (while its successor model is more relevant to current developments, it has identical architecture to the old one, and if you read about GPT-2 you will understand GPT-3).

Moreover, we have already been tackling the issue of neural nets on our own terms: the Objective Revision Evaluation Service has been running fine for several years. It seems to me that, if we were to approach these technologies with open minds, it could be possible to resolve some of our most stubborn problems, and bring ourselves into the future with style and aplomb. I mean, anything is possible. For all we know, the Signpost might start putting out print editions.

– J, AC, B, S

• Wikimedia Summit event report: Wikimedia Deutschland has published further reports on the 2022 Wikimedia Summit: the Event Report including financial info (overall cost was a little over 500,000 euros) and the participants' feedback report (the hybrid format appears to have been a success this time). Further documentation of the event is available here.
• Page views: Page views in September for all wikis have been up in relation to the past months with 23,657,615,038 views in comparison to 23,033,712,122 in August. Wikipedia has been growing in popularity since January of 2016, when page views were 20,865,413,322. According to Wikimedia Statistics, the number of page views seems to be growing steadily.
• Administrators: TheresNoTime voluntarily relinquished their Checkuser and Oversight permissions after remedies revoking these privileges in the "Reversal and reinstatement of Athaenara's block" arbitration case reached 5 to 1 in favour and also resigned as an admin; they remain a Steward and later reclaimed their admin status here on the English Wikipedia as well.
• Milestones: Chechen Wikipedia reached 500,000 articles; the traditional orthography Belarusian Wikipedia reached 80,000 articles; Banjar Wikipedia and Hakka Wikipedia reached 10,000 articles; and Dagbani Wikipedia, Guarani Wikipedia, and Maltese Wikipedia reached 5,000 articles. In Wiktionaries, Sango Wiktionary reached 40,000 entries; Shan Wiktionary reached 30,000 entries; and Lombard Wiktionary reached 20,000 entries.
• Articles for Improvement: This week's Article for Improvement is Medjed, followed by Human history (starting 5 December), then Telephone line (beginning 12 December). Please be bold in helping improve these articles!
• BernsteinBot disabled: HaleBot has been updating the List of Wikipedians by number of edits since mid-October. We thank BernsteinBot for its service.
• Easy, automatic journal access for editors: The Wikipedia Library has made Taylor & Francis academic journals automatically available for anyone with a Wikimedia unified login, using OCLC EZproxy (via T&F press release)