The Weekly Weird #65

Britain too sunny, fake accents, fake experts, fake sperm racing, deepfakes have a heart, TV shills for WorldID, AfD is officially 'extreme-right', UK supermarket FaRT, football FaRT

Hello again! Once more we gather at the base of Mad Mountain, to stare up at the steaming pile of dystopia dumped on our doorsteps by the previous seven days.

Before we get into it, here’s a quick clip of a Chinese crowd-control robot malfunctioning and lunging at the public…

Good times!

• Airbike Is Airwolf¹: The Polish company Volonaut have unveiled the AirBike, a “superbike for the skies” that can travel at up to 124mph. Described as “the obsession of its creator, entrepreneur and inventor Tomasz Patan”, it almost makes you excited to be living in the future.

• Reform Performs: Reform, the inheritor of the mantle of the Brexit Party, which itself was the party formerly known as the UK Independence Party (UKIP), has stomped to a resounding victory in England’s local elections (with 14 of 23 councils reported).

  

  Four years ago, it would have been considered unthinkable for a third party to deliver such barnstorming results. In fact, back in 2021, Reform’s leader Nigel Farage was moonlighting on the celebrity-for-hire website Cameo. Here he is delivering a well-lit and characteristically sincere birthday message to ‘Hugh Janus’.

  The best bit: He’s still on there. If you fancy a greeting from His Chinlessness, or want to make a contribution to hilarity by coming up with a script he can deliver for our shared amusement, his face and voice can be yours from £71.25.

  Reform have not only swept local councils but also secured a mayoral position for the first time in the party’s history. Dame Andrea Jenkyns became the Mayor-elect of Greater Lincolnshire with 42% of the vote, the news of which was shared by the BBC in an even-handed fair-minded way:

  

  Sadly, the fact that she was a Conservative Member of Parliament from 2015 to 2024 apparently didn’t make her a ‘former MP’ with a “knack for bouncing back” in the eyes of the always-objective BBC.

• AfD Is Officially ‘Extreme-Right’: The Alternativ für Deutschland, holder of 152 out of 630 seats in Germany’s Bundestag, “has been designated as right-wing extremist by the country's Federal Office for the Protection of the Constitution,” as per the BBC. The designation comes, according the agency’s statement, “due to the extremist character of the entire party, which disregards human dignity.”

  • “It aims to exclude certain sections of the population from equal participation in society, to expose them to non-constitutional unequal treatment and thus to assign them a legally devalued status. Specifically…German citizens with a history of migration from Muslim countries are not considered to be equivalent members of the German people ethnically defined by the party.”

  The finding comes just after the AfD placed second in the country’s federal elections with 20.8% of the vote, amid ongoing efforts to have the party banned.

  From the BBC:

  • “The deputy leader of the Social Democrat SPD, Serpil Midyatli, said it was now in black and white what everybody already knew. "It's clear for me that the ban has to come," she said, according to German press agency dpa.”

  Meanwhile, AfD leader Alice Weidel (pictured below demonstrating her preference that air-quotes be used when referring to her party as “extremist”) has been calling for closing Germany’s borders, withholding benefits from people without residence permits, and carrying out “remigration” on a grand scale².

  

• TV Shills For WorldID: A local San Francisco television station reported on the flagship store of Sam Altman’s Tools for Humanity Corporation, where The Orb can scan your iris in order to issue you with digital proof of you being a human. The anchor throws to USA Today’s tech-life columnist Jennifer Jolly on location with The Orb for an explanation of the technology with the cheery line “This looks like something my optometrist would use, Jen.” Jolly responds with an upbeat description of a dystopian hellscape:

  • “Well, hopefully it's something your optometrist will use, your bank will use, every store you walk into will use. That's the dream behind this…”

  The whole performance gives off retro infomercial vibes.

  With AI overwhelming the internet with slop, hallucinations, and garbage content, WorldID is being pitched as a way to “to fix the problem one eyeball scan at a time”. The fact that one of the major players in the AI space, OpenAI, is controlled by the same guy shilling the “solution” passes without comment.

  Jolly delivers the company’s stump speech on privacy, claiming that scanning your iris “doesn't capture any personal information” because “it instantly gets deleted,” leaving behind only “a digital barcode for your eyeball that's encrypted, like super encrypted, split into pieces and then scattered across a decentralized network.” Then it gets genuinely funny:

  • “I like to think of it kind of like a glitter bomb. So even if like somebody breaks in, if they get into it, like, can you imagine trying to piece together a glitter bomb? That's what these guys say this would be like.”

  Yeah, totally. A glitter bomb is a collection of tiny pieces that are all identical, so if you get one piece of glitter, you have the same thing as you would with any other piece of glitter. So either this analogy is terrible, or it’s apt and the data “scattered across a decentralized network” is multiplied rather than divided. Call me old-fashioned, but I’d suggest that the best way to protect your privacy is to not permit a billionaire’s creepy tech company to scan your iris. Besides, if someone manages to hack the “digital barcode for your eyeball”, how are you going to fix that breach?

  Tools for Humanity have announced pending partnerships with Tinder, OkCupid, Visa, and the gaming company Razor. As Jolly puts it, “this is only as good as the people they can get to adopt it, it needs mass adoption.” How amazing would that be?

  • “[I]magine if you go on Facebook or you go on Instagram, and for the first time in as long as you can remember, even YouTube, you know that what you're watching was created by a person. You know that the person you're engaging with is actually a real person.”

  The company is opening six more locations across the United States, and the timeline is reasonably tight. “I think we'll know within a year from now whether it will work or not,” Jolly tells the anchor. “But I think within two to three years this will be the new protocol for like your entry level engagement to doing anything on the internet.”

• Fake Sperm Racing: Sperm Racing, the imaginatively-titled company organising what it claims (erroneously, it turns out) is a “world’s first”, managed to raise $1 million in seed capital, get a partnership with the online gambling site prediction market Polymarket, hold a live event in which two bros knock one out and put their swimmers to the test, and then got rumbled for allegedly faking the racing part, no mean feat in the space of a few months.

  In The World’s First Sperm Race Seemed Too Good to Be True. It Was. the Free Press reported on questionable ethics related to the ‘sport’, namely that the race itself appears to have been computer-generated, meaning that the outcome was known to some people while bets were still being placed by the public.

  • “As the Sperm Racing crew prepared for the main event, [FP reporter] Austyn made his way backstage where he peered into the control room. He saw that prerecorded clips of the sperm race had been uploaded into the “live” video stream. That’s not all: They had clearly been labeled with the name of the winner of each round: Two clips read “Tristan Win,” and one “Asher Win.”

    The “live” race hadn’t even begun, and Austyn knew the winner.

    “I could have gone onto Polymarket and placed a bet right then, knowing exactly who won—and so could anybody on the Sperm Racing crew,” Austyn told me.”

  When questioned about the timing, Eric Zhu, the 17-year-old behind the new sporting phenomenon, admitted that “they actually did this an hour before” and that the race, as shown to the public, “use[d] computer vision to basically trace the sperm and then overlay [them] onto a 3D model.”

  The first clue that something was amiss should have been when the races took only a few minutes. As reported by the Daily Mail before the event: “Sperm typically swim at about 5mm per minute, meaning each race will take at least 40 minutes.”

  So two guys cracked one out, their sperm was put through a microscopic race track and recorded, the video was mapped into a 3D environment to mimic what actually happened, and the results were known in advance and re-ordered “to make it more interesting.” All this is to say that, as per their post on X, Sperm Racing are still insisting their sport is real in the ways that matter, and declare that they have “the goal of one day seeing it in the Olympics.”

  If you’ve got sperm to spare, their website even has a ‘Tryouts’ page where, after submitting a sample, “you'll be placed onto the Sperm Racing Amateur Leaderboards — where we scout the next generation of professional athletes” and “one day, racing sperm could be your full-time career.”

  Dare to dream.

  

• Fake Accents: “The world’s biggest call centre company is using artificial intelligence to “neutralise” Indian accents for Western customers”, reports the Telegraph. Teleperformance, a French company, is “applying real-time AI software on phone calls” supplied by the American company Sanas. AI ‘accent conversion’ is a burgeoning business, with other companies like Krisp also stepping into the space.

  Here’s an old demo of the Sanas software:

  From The Telegraph:

  • “Sanas has been criticised in the past for making people’s voices “sound whiter”. The company says that call centre workers like the technology and it leads to more opportunities in countries such as India and The Philippines.”

  For some reason, this is what came to mind:

  There’s already a growing problem of scammers cloning voices to help them take advantage of people, as one two-time victim explained in a YouTube video. As the technical toolkit grows for businesses, it unfortunately grows just as quickly for people looking for a criminal or fraudulent angle.

  In January, CNN reported that the Prime Minister of Thailand received a scam call featuring the AI-generated voice of an unnamed world leader.

  • “Paetongtarn Shinawatra did not reveal who the computer was mimicking, but said she received a message in a voice identical to a well-known leader.

    “The voice was very clear, and I recognized it immediately. They first sent a voice clip, saying something like, ‘How are you? I want to work together,’ and so on,” Paetongtarn said.”

  In February, Reuters reported on another AI voice scam that defrauded a well-meaning Italian businessman out of nearly a million euros.

  • “Fraudsters used AI to mimic the voice of Italian Defence Minister Guido Crosetto, making calls that claimed to seek urgent financial assistance for the release of kidnapped Italian journalists in the Middle East.”

  It’s one thing for a prime minister to get a call soliciting a donation, or for a cabinet minister to get cloned, but when certain politicians already engage in ridiculous behaviour like selling Bibles, NFTs, watches, or cryptocurrency while running for or even serving in office, it doesn’t seem paranoid to be concerned that a scammer will eventually come up with something so believable to their mark that it will pass the smell test and end up causing a serious geopolitical incident.

  In unrelated news, the Telegraph just reported that the “Emirati investment firm MGX, which is backed by Abu Dhabi’s sovereign wealth fund” is “planning to spend $2bn (£1.5bn) on one of Donald Trump’s cryptocurrencies.”

• Fake Experts: Rob Waugh at the Press Gazette broke a story last month on AI comment-farming, in which journalists seeking comment are fed artificially-generated content and, in at least one case, quoted an ‘expert’ who didn’t exist. Even when there appears to be a person involved, the shadiness and caginess led them to conclude that there was no evidence of ‘expert’ status, as in the curious case of Barbara Santini:

  • “For years, her commentary has appeared in national and international publications including Vogue, Metro, Cosmopolitan, The i Paper, the Express, Hello, Shape, Cosmopolitan, Women’s Health, Yahoo, Good Housekeeping, the Telegraph, the Daily Star, the Daily Mail and The Sun. In terms of media mentions, Santini is probably one of the most prominent psychologists in the UK.”

  • “Via email, Santini said that she was “unable to take calls”. Asked to provide proof of who she was in the form of qualifications or details about her practice, she did not respond.”

  Santini threatened legal action against the Press Gazette, unlike “Rebecca Leigh”, who the Gazette concluded “is not real”, and about whom a website showing her as a contributing writer stated for the record “the name and the photo are not real, as we provide our writers with anonymity guarantees.”

  In a follow-up article, Waugh reported that “[d]ozens of stories have been deleted and amended by leading publishers” after the Press Gazette investigation.

  • “ResponseSource, the journalist request service which one alleged fake profile used to send comments to dozens of journalists in the UK, has instructed staff to be “vigilant”. The network also explained the “blind spot” which allowed a fake profile to be featured dozens of times in national newspapers over a number of years.

    The Sun has confirmed that it will be searching for and removing all articles involving one apparently fake expert who had commented on issues for the paper ranging from what the colour of one’s underwear says about your personality to the intricacies of the “Italian chandelier” sex position.

    Publisher Reach is now removing a number of articles across its publications, which include the Mirror and Express titles. The Telegraph is understood to have warned freelances about the pitfalls of using journalist-request services. Yahoo News has either deleted or amended more than a dozen articles.

    Press Gazette has compiled a dossier of more than 100 articles which have appeared in Mail Online, The Sun, Mirror, Express, HuffPost UK, Yahoo News, Metro, The Independent and the Telegraph based partly, or entirely, on input from apparently fake experts.”

  Part of the problem might be that journalists, underfunded and pressed for time, use comment-seeking websites to get ‘experts’ to add colour or background to their work rather than actually looking for specific people who they want to involve in the piece they’re working on. More broadly, bullshit has always been a factor in media, and AI is just making it more common and harder to spot. There’s also an element of trajectory reading to be done here, if we look at what the problem is being diagnosed as and what solutions are getting airtime.

  As the futurist Tracey Follows told the Gazette: “We’ll probably get to an online world where everybody is required to have much more of a digital identification, some verification, that not only are you a person – but you are this person or that person.”

  Okay, everybody line up for The Orb.

• Football FaRT: Danish man-kick-ball team FC Copenhagen have received permission to FaRT all over their fans. Biometric Update reports that the Danish Data Protection Agency has given the club permission for “processing of biometric data for the purpose of uniquely identifying a person using automatic facial recognition when holding any event at Parken and for FC Copenhagen’s away matches at stadiums other than Parken Stadium.” The technology “will be used for biometric access control at stadium entrances, mobile facilities, training matches and other events "in connection with the holding of FC Copenhagen football matches."” The permission covers the club, not the team’s stadium, so the data gathering can only take place for football matches, not in a blanket manner at all events.

• UK Supermarket FaRT: British supermarket Asda is trialling facial recognition at five stores in Greater Manchester to combat “an epidemic of retail crime”, according to their press release. “The technology is integrated into Asda’s existing CCTV network and works by scanning images and comparing the results to a known list of individuals who have previously committed criminal activity on an Asda site.”

• Deepfakes Have A Heart: In news to warm even the coldest cockles, researchers have published a new study showing conclusively that deepfake videos of faces can now include “[m]inute changes in skin color due to blood flow in time with heartbeats [that] have been used to differentiate deepfakes from real videos in the past”, as per Biometric Update.

  • “Our high-quality deepfakes exhibit valid heart rates and their rPPG³ signals show a significant correlation with the corresponding driver video that was used to generate them. Furthermore, we show that this also holds for deepfakes from a publicly available dataset.

    Previous research assumed that the subtle heart-beat-related signals get lost during the deepfake generation process, making them useful for deepfake detection. However, this paper shows that this assumption is no longer valid for current deepfake methods.”

Britain Too Sunny

With their proverbial finger firmly on the pulse of the nation, the citizenry of which frequently and loudly complain that the UK is too hot with not enough cloud cover, the British government has officially greenlit a geoengineering/solar radiation modification (SRM) project intended to “dim the sun” by blocking sunlight to mitigate the effects of climate change.

The Advanced Research & Invention Agency (ARIA) calls it Future Proofing Our Climate and Weather, and the programme has been granted £50 million in funding for “outdoor geoengineering experiments”, as reported by the Guardian. “The programme, along with another £11m project,” they wrote, “will make the UK one of the biggest funders of geoengineering research in the world.”

Geoengineering is controversial and some previous planned outdoor experiments have been cancelled after strong opposition. Most geoengineering proposals aim to block sunlight reaching the Earth’s surface, for example by launching clouds of reflective particles into the atmosphere or using seawater sprays to make clouds brighter.

From The Independent:

Another method involves thinning natural cirrus clouds, which act as heat-trapping blankets.

If successful, less sunlight will reach the earth’s surface and in turn temporarily cool the surface of earth.

It’s thought to be a relatively cheap way to cool the planet, but critics have warned it could cause catastrophic disruption to weather patterns and even shift rain from areas that are vital for food production.

The Programme Director at ARIA, Professor Mark Symes, assured the Guardian that “no toxic substances would be released, an environmental impact assessment would be published before outdoor experiments and…local communities would be consulted”.

The University of Cambridge’s Centre for Climate Repair was keen not to let Symes and ARIA hog the spotlight:

These outdoor experiments won’t stand alone. At the start of April, we announced new funding to support our modelling research, which will use Earth Systems Models to evaluate the potential risks and effects of different solar radiation management (SRM) techniques. These are the same kinds of models used to predict climate change and weather effects, and they can help build up a picture of the large-scale impacts to complement the real-world experiments.

ARIA’s initial release about the programme was titled Exploring Options for Actively Cooling the Earth, and its thesis makes clear that “in the absence of substantial physical (as opposed to simulated) data on the mechanisms behind how these concepts might work (and what their effects might be), there is no prospect of being able to make proper judgements on what are or are not feasible, scalable, and controllable approaches for cooling the Earth.”

[W]e plan to fund not only the experiments themselves, but also the necessary modelling, simulation, observation and monitoring required to support the experiments, as well as research into the ethical, governance, law, and geopolitical dimensions of the approaches under investigation.

The “risk” of reaching “tipping points” such as the loss of Arctic ice in winter or “dieback” in the Amazon is given as a key motivation for the programme.

The thresholds for many such tipping points remain far from clear, but it seems likely that a certain amount of continued global warming is already locked in, even with rapid decarbonisation, on account of the amount of carbon dioxide already in the atmosphere…

The questions at the heart of the programme are set out in the introduction.

How (or whether) research into methods that could be used to reduce global temperatures should be conducted throws up a number of open questions. For example, do the risks of unintended consequences and moral hazard associated with learning more about Earth-cooling approaches outweigh the risks of continued global warming without researching any intervention strategies? How should we weigh the risks associated with researching approaches for reducing global temperatures against the risk that the world discovers in 2040 or 2050 that efforts to achieve net zero and to remove carbon dioxide from the atmosphere have been insufficient to prevent very detrimental tipping points? In such a scenario, what might be the risks of hurried deployment of under-researched climate engineering approaches where we have little understanding of the consequences? And if these approaches were deployed, what might be the risks associated with a later, more sudden deployment relative to a slower ramp-up? Or the risks of termination shocks if deployment were suddenly stopped?

The upshot is that “ARIA has come to the view that the risks of not being able to answer such questions are greater than the risks of researching approaches for actively cooling the Earth through a well-governed research programme.”

The subsequent release, Exploring Climate Cooling, went out in September 2024.

Computer modelling and indoor testing are essential and necessary first steps in establishing the basic science behind how (or whether) a particular approach might work. However, modelling and indoor testing alone cannot provide all the data necessary to predict the effects of a given approach on the real world with a suitable level of confidence. Controlled outdoor experiments are therefore likely to be required to truly advance our understanding of the phenomena underlying potential approaches. The overarching goal of this programme is to answer the most critical technical and fundamental questions on the practicality, measurability, controllability, and likely (side-)effects of approaches that might one day be used to actively cool the Earth.

This isn’t the first or only programme of its kind - as written about in Weirds past, 2024 saw a “secret climate experiment” carried out in San Francisco, “the country’s first outdoor experiment to limit global warming by increasing cloud cover to reflect more sunlight away from the Earth.” As reported in Politico, “[t]he Coastal Atmospheric Aerosol Research and Engagement, or CAARE, project, led by researchers at the University of Washington, kept a tight lid on the project details...out of concern "that critics would try to stop them"”.

In 2023, the Climate Overshoot Commission, who could not possibly be accused of objecting to dealing with climate change, published a report on climate change mitigation and stated that although “[s]olar radiation modification should be researched, and its governance discussed”, “[c]ountries should adopt a moratorium on the deployment of solar radiation modification and large-scale outdoor experiments that would carry risk of significant transboundary harm”.

“Transboundary harm” sounds like college talk for unintended consequences, which are the human race’s stock in trade.

And of course, as always, The Simpsons already did it.

---

That’s it for this week’s Weird, everyone. Thanks as always for reading.

Outro music is Longview by Green Day, perhaps the most commercially successful song about “relaxing in a gentleman’s way”, dedicated to all the future ‘fertility athletes’ out there.

Stay sane, friends.

1. This video of Ernest Cline’s slam poem Airwolf will explain.

2. This clip from one of her speeches is in German. You can activate auto-translate on YouTube to get English subtitles.

3. Remote photoplethysmography