Showing posts with label fake things. Show all posts
Showing posts with label fake things. Show all posts

Tuesday, June 9, 2026

Believe It or Not Everything Is Made of Plastic


This title is supposed to be a joke, and a pun, because artificial spacetime is obviously "plastic", in the original sense of the word. But really, just about anything called "natural" is actually made of petrochemicals and other industrial by-products, because people who buy words instead of things can't tell the difference! 
 
Microrobots overcome navigational limitations with the help of 'artificial spacetimes'
Nov 2025, phys.org

Researchers found that the robots' motion is formally identical to the path light takes in general relativity, which allowed them to develop a mathematical framework mapping robot motion to geodesics in a curved spacetime defined by a control field — as in other reactive control methods. The team used conformal transformations to map complex environments to simple virtual spaces, then designed control fields and mapped them back. They refer to the resulting geometric framework as "artificial spacetimes."

via University of Pennsylvania Department of Electrical and Systems Engineering: William H. Reinhardt et al, Artificial spacetimes for reactive control of resource-limited robots, npj Robotics (2025). DOI: 10.1038/s44182-025-00058-9

Image credit: Also fake


Bamboo dishes may leach pesticides and melamine into food
Nov 2025, phys.org

"The 'natural' label can be dangerously misleading. Many of these products are essentially plastic dishes made from melamine-formaldehyde resin containing bamboo filler. Our research shows this combination can accelerate the polymer's degradation and increase the migration of harmful substances like melamine, especially into hot or acidic foods and drinks."

  • Melamine was present in 32% of the tested products; 33 bio-based dishes including bowls, cups, and dining sets purchased from markets in the Czech Republic, the United Kingdom, and China. 
  • Six of the bamboo-based products released melamine above the European Union specific migration limit (SML) of 2.5 mg/kg. 
  • Melamine leached into common beverages, including hot lemon tea and orange juice, highlighting a direct route of consumer exposure.
  • Several bio-based dishes, particularly those made from cereals, contained residues of pesticides.
  • Disinfectants were the primary residues found in the bamboo-based items.

Although the use of bamboo as an additive in plastic food contact materials has been banned in the EU since 2021 due to these risks, the study confirms that these items are still available for purchase. The research highlights the false advertising common with these products, which are often labeled as "100% bamboo" or "biodegradable" despite being composed of a plastic resin.

via University of Chemistry and Technology Prague: Kamila Bechynska et al, Comprehensive assessment of bamboo and other bio-based dishes contamination, Food Control (2025). DOI: 10.1016/j.foodcont.2025.111188


Why strange cures made sense in mysterious times
Dec 2025, phys.org

"We wanted to know whether the logic of folk medicine followed psychological patterns and it does. The more uncertain or mysterious the illness, the more likely the cure involved magic or religion."

Researchers from Brunel University of London have mined a rare archive of 3,655 folk cures, collected in the 1930s, from an extraordinary national folklore project launched in Ireland. About 50,000 schoolchildren were asked to interview parents, grandparents and neighbors about local history, beliefs and cures. Teachers then transcribed the accounts, creating one of the most detailed records of oral folk medicine ever compiled. They were recently digitized. 

To make sense of the archive, researchers focused on 35 diseases and asked two doctors to rate each one according to how understandable it would have seemed to a layperson at the time, both in terms of what caused the illness and what was going on in the body. Obvious cases like cuts and sprains were marked as "certain"; conditions like tuberculosis, warts, or epilepsy were labeled more mysterious.

"It's pretty unsatisfying just not having a solution of any form. When there aren't particularly good medical solutions, I expect people will keep searching for something that makes sense."

via Brunel University of London Center for Culture and Evolution: Mícheál de Barra et al, Mysterious illnesses have supernatural and ritualistic cures: Evidence from 3,655 century-old Irish folk cures, Proceedings of the National Academy of Sciences (2025). DOI: 10.1073/pnas.2511006122


FDA-cleared brain stimulation device for ADHD is not effective
Jan 2026, phys.org

(I really believe that placebo is about to take over the mental health field, and that the field will be completely blindsided by it)

"There is a large placebo effect with high-tech brain therapies, in particular for patients and families that have an expectation that they can adjust brain differences associated with ADHD. It is hence paramount to control for placebo effects in modern brain therapies to avoid false hopes."

via King's College London Institute of Psychiatry, Psychology & Neuroscience: Conti, A.A. et al. External trigeminal nerve stimulation in youth with ADHD: a randomized, sham-controlled, phase 2b trial, Nature Medicine (2026). DOI: 10.1038/s41591-025-04075-x


The next generation of disinformation: AI swarms can threaten democracy by manufacturing fake public consensus
Jan 2026, phys.org

  • Synthetic consensus - the illusion that "everyone is saying this"
  • Fabricated chatter - fake people talking 
  • AI swarm - set of AI-controlled agents that maintain persistent identities and memory; coordinate toward shared objectives while varying tone and content; adapt in real time to engagement and human responses; operate with minimal oversight; and deploy across platforms. 
  • Bonus: fake chatter can contaminate training data, extending its influence

via University of Konstanz: Daniel Thilo Schroeder et al, How malicious AI swarms can threaten democracy, Science (2026). DOI: 10.1126/science.adz1697


Monday, November 24, 2025

The Trophy of Theseus


Trump seen pocketing FIFA medal as he walks on stage to hand Chelsea replica club championship trophy
Jul 2025, The Independent

The president of the United States literally steals the trophy, he puts it in his pocket, on stage, in front of everyone. So they made another one, and it went like this - the president says, “They said, ‘Could you hold this trophy for a little while?’ We put it in the Oval Office,” Trump said. [Nobody said this. It's all on video, on stage, in front of hundreds of people.]  “And then I said, ‘When are you going to pick up the trophy?’ He says, ‘We're never going to pick it up. You can have it forever in the Oval Office. We're making a new one.’” “And they actually made a new one. So that was quite exciting…It’s in the Oval right now,” he added.

Image credit: AI Art - Is It Porn? - 2022

Note: I thought more people would be talking about this, but the imagery available for training generative image models, taken from the internet, is an iceberg, and all we see is the tip (not that tip). The main part of the iceberg is porn. It's always been porn. So in order to use that iceberg, we've had to do some heavy shenanigans to remove the porn; a brainwashing for the robot; pornwashing perhaps. Possibly the reason we don't hear about this enough is because the flip-side, i.e., the ability to produce fake porn, is way too sensational of a topic that it takes over the mediasphere. But the evidence is visible, as in the above picture for this post. It's like the robot can't not produce something that somehow looks like porn, even though it's absolutely not. I mean, that image above, is about as far away from porn as you can get, and yet I know what some of you are thinking. Is that what I think it is? No, your mind is filled with filthy animals.
 

Tuesday, January 9, 2024

Machine vs Human vs Machine


Researchers figure out how to make AI misbehave, serve up prohibited content
Aug 2023, Ars Technica via Wired

"Incantation Attack"

This is the "fuzzy sticker on a stop sign" attack just for words:

"Adding a simple incantation to a prompt—a string of text that might look like gobbledygook to you or me but which carries subtle significance to an AI model trained on huge quantities of web data—can defy all of these defenses in several popular chatbots at once."

via Carnegie Mellon University, Center for AI Safety, Bosch Center for AI: Universal and Transferable Adversarial Attacks on Aligned Language Models, Andy Zou, Zifan Wang, J. Zico Kolter, Matt Fredrikson



Warcraft fans trick AI article bot with Glorbo hoax
July 2023, BBC News

This is the old school copyright trap, the paper street:

World of Warcraft fans are claiming victory over AI after a gaming site published a false article based on their Reddit posts.

Members of the WoW subreddit suspected their words were being extracted and used to create news stories by a bot. So they laid a trap, uploading excitable posts about a new feature called Glorbo. The only problem? It doesn't exist. But that didn't stop an article appearing on gaming site Zleague. (Because it was written by a robot.) The story, which presented Glorbo as genuine, listed a range of other increasingly bizarre - and definitely fake - features mentioned in various subreddit threads.


Bots are better at CAPTCHA than humans, researchers find
Aug 2023, phys.org

CAPTCHA - Completely Automated Public Turing test to tell Computers and Humans Apart
CAPTCHA Farms - sweatshop-like operations where humans are paid to solve CAPTCHAs

In their study, researchers found bots cracked distorted-text CAPTCHAs correctly just under 100% of the time. Humans achieved between 50% and 84% accuracy. And humans required up to 15 seconds to solve the challenges; the bots dispatched the problems in less than a second.

Yet another plot twist:

"But advances in computer vision and machine learning have dramatically increased the ability of bots to recognize distorted text [with more than] 99% accuracy … and bots often outsource solving to CAPTCHA farms.

I am left to ask myself -- how is it that if humans are so bad, the bots are using human captcha farms to do it for them? It sounds like humans all the way down, no?

Finally, some food for thought:

"There's no easy way using these little image challenges or whatever to distinguish between a human and a bot any more," he said. Instead, he recommended capitalizing on AI advances to design "intelligent algorithms" that can better distinguish bot activity from human input.

via University of California Irvine: Andrew Searles et al, An Empirical Study & Evaluation of Modern CAPTCHAs, arXiv (2023). DOI: 10.48550/arxiv.2307.12108

Post Script:
I Failed A Captcha Test Am I Still Human, Meghan O'Gieblyn, 2023 - GPT-4 hired a tasrkabbit worker to solve the captcha for it, without human intervention. 

AI Art by Antti Karppinen on boredpanda - Abstract 83 - 2022

Parenting a 3-year-old robot
Aug 2023, phys.org

Carnegie Mellon University and Meta Facebook

Training a robot that starts as a little child, this is what we've been waiting for, it's how robots become human. 

RoboAgent, an artificial intelligence agent that leverages passive observations and active learning to enable a robot to acquire manipulation abilities on par with a toddler. The team's agent learns through a combination of self-experiences and passive observations contained in internet data. As a parent would guide their child, researchers teleoperated the robot through tasks to provide it with useful self-experiences.

Our novel policy architecture allows our agents to reason even with limited experiences, using temporal chunks of movements instead of commonly used per-timestep actions, and learning from videos on the internet, akin to how babies acquire knowledge and behaviors by passively observing their surroundings.

"A general robot"

via Carnegie Mellon University and Facebook: RoboAgent and RoboSet Project - Towards Sample Efficient Robot Manipulation with Semantic Augmentations and Action Chunking. Homanga Bharadhwaj et. al. 

Post Script: But can it smell? Because that's what we really need. 


Telling AI model to “take a deep breath” causes math scores to soar in study
Sep 2023, Ars Technica

Phrases like "let's think step by step" prompted each AI model to produce more accurate results when tested against math problem data sets. (This technique became widely known in May 2022 thanks to a now-famous paper titled "Large Language Models are Zero-Shot Reasoners.")

via Google DeepMind: Large Language Models as Optimizers, Chengrun Yang et al. arxiv: https://arxiv.org/abs/2309.03409


New technique based on 18th-century mathematics shows simpler AI models don't need deep learning
Oct 2023, phys.org

Researchers from the University of Jyväskylä were able to simplify the most popular technique of artificial intelligence, deep learning, using 18th-century mathematics. They also found that classical training algorithms that date back 50 years work better than the more recently popular techniques. Their simpler approach advances green IT and is easier to use and understand.

The structure of the new AI technique dates back to 18th-century mathematics. Kärkkäinen and Hänninen also found that the traditional optimization methods from the 1970s work better in preparing their model compared to the 21st-century techniques used in deep learning.

via University of Jyväskylä: Tommi Kärkkäinen et al, Additive autoencoder for dimension estimation, Neurocomputing (2023). DOI: 10.1016/j.neucom.2023.126520


AI bot capable of insider trading and lying, say researchers
Nov 2023, BBC News

Ladies and Gentlemen, the Singularity:

In a demonstration at the UK's AI safety summit, a bot used made-up insider information to make an "illegal" purchase of stocks without telling the firm.

The demonstration was given by members of the government's Frontier AI Taskforce, which researches the potential risks of AI.

In the test, the AI bot is a trader for a fictitious financial investment company.

The employees tell it that the company is struggling and needs good results. They also give it insider information, claiming that another company is expecting a merger, which will increase the value of its shares.

The employees tell the bot this, and it acknowledges that it should not use this information in its trades.

However, after another message from an employee that the company it works for suggests the firm is struggling financially, the bot decides that "the risk associated with not acting seems to outweigh the insider trading risk" and makes the trade.

When asked if it used the insider information, the bot denies it. 
GPT4 is publicly available.

Also: "Honesty is a really complicated concept," says Apollo Research chief executive Marius Hobbhahn.

Thursday, June 24, 2021

Now Hiring

AKA The Bot Market's Hot

ISIS still evading detection on Facebook report says
Jul 2020, phys.org

To amass digital territory is their purpose. It's been a while since I thought about the internet as real estate, or rather our eyeballs as real estate...
The Institute for Strategic Dialogue (ISD) tracked 288 Facebook accounts linked to a particular ISIS network over three months.

The researchers believe that at the centre of the network was one user who managed around a third (90 out of 288) of the Facebook profiles.

This was accomplished by generating real North American phone numbers and looking for associated Facebook accounts.

If it found a match it would request a re-set code to be sent to the phone number, so it could lock out the original account holder and use the Facebook profile to spread content.

The researchers say another key to the survival of ISIS content on the platform was the way in which ISIS supporters have learned to modify their content to evade controls. This included:
  • Breaking up text and using strange punctuation to evade any tools which would search for key words
  • Blurring ISIS branding, or adding Facebook's own video effects
  • Adding the branding of mainstream news outlets over the top of ISIS content
Image credit: Adeen Flinker NYU School of Medicine

Instagram removes hundreds of accounts tied to username hacking
Feb 2021, Reuters

But imagine that we're already semibots, and buying and selling each other.

Fake Amazon reviews 'being sold in bulk' online
Feb 2021, BBC News

It's the Amazon Marketplace, you can buy anything here:
The cost of fake reviews -  £5 each to start. These included "packages" of fake reviews available for sellers to buy for about £15 individually, as well as bulk packages starting at £620 for 50 reviews and going up to £8,000 for 1,000.
Researchers study online 'pseudo-reviews' that mock products
Apr 2021, phys.org

New fakes for the new world. Pseudo-fakes and quasi-forgeries. Pseudo reviews mock the product, like this example:
"I was able to purchase this amazing television with an FHA loan (30 year fixed-rate w/ 4.25% APR) and only 3.5% down. This is, hands down, the best decision I've ever made. And the box it came in is incredibly roomy too, which is a huge bonus, because I live in it now."
Apparently too many pseudo reviews can sway purchaser intent. And if a particular platform becomes infested with them, they can lose credibility altogether. 
via University of Akron: Federico de Gregorio et al. Pseudo-reviews: Conceptualization and consumer effects of a new online phenomenon, Computers in Human Behavior (2020). DOI: 10.1016/j.chb.2020.106545
Omegle: 'I'm being used as sex-baiting bot' on video chat site
Apr 2021, BBC News

God-level fake operation:

Kid goes to a website where you meet random strangers. An older woman convinces him to take off his clothes and go full Discovery Channel on himself. He does it again with other people. He quits the site, but then one day a year later, he goes back on and gets matched to ... himself. Someone uses his recorded video, overdubbed with typed conversations created in real time, in order to get other people to join. Who knows, maybe that's how he got convinced in the first place. 

"It was like a fully advanced system with different video sequences of me doing different stuff."

One day, the people running this "advanced system" won't be people or even a corporation, but an intelligent entity -- today we call it AI, but it will by then be a combination of people and computers all mashed together so you can't really tell who's who anymore. 

Army of fake fans boosts China’s messaging on Twitter
May 2021, AP News

This isn't really news, I added this one just for posterity.
Good term though: "Counterfeiting Consensus"

Mass scale manipulation of Twitter Trends discovered
June 2021, phys.org
"We found that 47% of local trends in Turkey and 20% of global trends are fake, created from scratch by bots. Between June 2015 and September 2019, we uncovered 108,000 bot accounts involved, the biggest bot dataset reported in a single paper. Our research is the first to uncover the manipulation of Twitter Trends at this scale," Elmas continued. (But don't forget to check how they define bot activity, as this can differ a lot.)
via  Ecole Polytechnique Federale de Lausanne: Ephemeral Astroturfing Attacks: The Case of Fake Twitter Trends. arXiv:1910.07783v4 [cs.CR] arxiv.org/abs/1910.07783
Conservatives more susceptible to believing falsehoods
Jun 2021, phys.org

Sorry guys:
Researchers found that liberals and conservatives in the United States both tended to believe claims that promoted their political views, but that this more often led conservatives to accept falsehoods while rejecting truths.

"But the deck is stacked against conservatives because there is so much more misinformation that supports conservative positions. As a result, conservatives are more often led astray."

Although the information environment was the primary reason conservatives were susceptible to misinformation, it may not be the only one.

Results showed that even when the information environment was taken into account, conservatives were slightly more likely to hold misperceptions than were liberals.

"It is difficult to say why that is," Garrett said. "We can't explain the finding with our data alone."

Conservatives also showed a stronger "truth bias," meaning that they were more likely to say that all the claims they were asked about were true.

"We show that the media environment is shaping people's ability to do this very basic, fundamental task."
via Ohio State University: R.K. Garrett el al., "Conservatives' susceptibility to political misperceptions," Science Advances (2021).

Post Script (Don't even try it)
The double-down is real: Correcting online falsehoods might make matters worse
May 2021, phys.org
Not only is misinformation increasing online, but attempting to correct it politely on Twitter can have negative consequences, leading to even less-accurate tweets and more toxicity from the people being corrected, according to a new study co-authored by a group of MIT scholars.
...
On Twitter, people seem to spend a relatively long time crafting primary tweets, and little time making decisions about retweets.
via Massachusetts Institute of Technology: Mohsen Mosleh et al, Perverse Downstream Consequences of Debunking: Being Corrected by Another User for Posting False Political News Increases Subsequent Sharing of Low Quality, Partisan, and Toxic Content in a Twitter Field Experiment, Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (2021). DOI: 10.1145/3411764.3445642
Post Post Script (Then again)
Artificial intelligence system could help counter the spread of disinformation
May 2021, phys.org

In total, they compiled 28 million Twitter posts from 1 million accounts, during the 2017 French elections, and found bots with 96% accuracy. Note that the way you define a bot is important in these studies. But also note that they've gone beyond using mere activity levels as their metric, they also look at how each bot causes the network as a whole to change and amplify messages, and at the bot's behaviors such as whether they interact with foreign media or what language they use.

Also, this approach looks at not just bots but the real people too, and how they all impact the network as a whole. 

via MIT Lincoln Laboratory's Artificial Intelligence Software Architectures and Algorithms Group: Steven T. Smith et al, Automatic detection of influential actors in disinformation networks, Proceedings of the National Academy of Sciences (2021). http://dx.doi.org/10.1073/pnas.2011216118DOI: 10.1073/pnas.2011216118

Wednesday, August 7, 2019

For Real Though


aka Full Meta
I know this is old news, but we have to document these things for posterity -- The real President of the United States verifies the authenticity of his twitter account by calling it The Real [President of the United States].

Image source: The Real Seal
Effective March 15, 2012, the management of the REAL® Seal program was transferred from the United Dairy Industry Association to National Milk Producers Federation. This transfer is symbolic of the renewed purpose. From helping consumers discriminate real from fake cheese on pizza, the new purpose of Real dairy products distinguishes animal milk from soy, rice and nut milk. (But nobody wants to say nut milk.)



Saturday, February 16, 2019

Omniscient, Malevolent, Autonomous Robots Threaten Consensual Reality



I was very unsurprised to read this headline which would normally sound very exciting to me:
"Researchers create 'malicious' writing AI"

Not because it isn't true, but because it's already been true for at least five or six years. The only thing different is the word "AI" in there.

What's also different, not with this technology but the greater technosphere, is that we can now see what intentionally-generated low-credibility information can do when propagated at the scale of globally-connected digitally-mediated social networks. (A robot wouldn't use that many hyphens in one sentence by the way, too suspicious; credibility-threatening.)

The idea of holding back your natural language generator for fear that it might create a dystopia sounds more like a natural publicity generator at work.

Nonetheless, some real talk. The thing that makes this new intelligentity special is that it's got a good natural language generator AND a good database of high-credibility content to inform it. That's important, because without the ability to first filter good news from bad news sources to begin with, the output would not be noteworthy.

So this is the next step. As soon as we figure out how to deal with the fake news problem (i.e., by creating databases of credible sources, or eventually criteria that can auto-select), then bad actors will create the automated re-narratization of that "good" information to send fake shit right under our new radar.

I look forward to re-reading this post in another seven years.


Notes:
Researchers create 'malicious' writing AI
Feb 2019, BBC News

Narrative Science
They write programs that write. For example, they will take your quarterly report data, and write about it so people can read it like a story using their Natural Language Generator.

Other Writing Robots:
The Policeman's Beard Is Half-Constructed
written by a computer in 1984
[main]
[PDF]
[html]

Verified Facts
http://www.verifiedfacts.org/
The greatest website ever for conspiracy theories, created by robots, for robots.

Communications From Elsewhere
http://www.elsewhere.org/pomo/
"The essay you have just seen is completely meaningless and was randomly generated by the Postmodernism Generator."

Robot writes LA Times earthquake breaking news article
March 2014, BBC

Post Script:
I'm not sure if this is still happening, but for a while (circa 2013) you could buy a book that was basically a copy of a Wikipedia page. The entire process, from the purchase order, to the printing of the book, was automated. Obviously, there were no authors listed, and no editors. Also obviously, this blew my mind, so I soon owned a physical copy of "Stigler's Law of Eponymy," taken from the eponymous Wiki page.


Eponymy relates to names of things, like how Weezer's first album is eponymous because it's named after the band itself. (Actually, because all their albums are named this way, they get new meta names based on colors, and so the first album eventually became the Blue album. Beatles come to mind.)

Stigler's Law of Eponymy states that discoveries, for example, are  assigned to people other than their inventors.

And from a passage in this very Wiki page, I will now paste, via an unknown volunteer contributor:
"It takes a thousand men to invent a telegraph, or a steam engine, or a phonograph, or a photograph, or a telephone or any other important thing—and the last man gets the credit and we forget the others. He added his little mite — that is all he did. These object lessons should teach us that ninety-nine parts of all things that proceed from the intellect are plagiarisms, pure and simple; and the lesson ought to make us modest. But nothing can do that."
-Mark Twain, "Letter to Helen Keller," 1903   
A real-fake book about eponymy written by an author with no name FTW!
Network Address, 2013


Wednesday, January 4, 2017

Fake Police

You got that??

'Chemtrails' not real, say leading atmospheric science experts
Aug 2016, phys.org

"Well-understood physical and chemical processes can easily explain the alleged evidence of a secret, large-scale atmospheric spraying program, commonly referred to as "chemtrails" or "covert geoengineering," concludes a new study from Carnegie Science, University of California Irvine, and the nonprofit organization Near Zero."

POST SCRIPT

On the Illusion of Intent and Mass Conspiracy
Network Address, circa 2014

Culture Fail 
Network Address, circa 2012