Post

How to get URL link on X (Twitter) App

On the Twitter thread, click on or icon on the bottom
Click again on or Share Via icon
Click on Copy Link to Tweet
Paste it above and click "Unroll Thread"!
More info at Twitter Help

Conspirador Norteño

@conspirator0

Jul 24, 2022 • 11 tweets • 8 min read • Read on X

Scrolly

@ZellaQuixote

It's a great day for a quick look at a variety of suspicious Twitter accounts with GAN-generated faces. #SundayShenaniGANs

(GAN = "generative adversarial network", the technology used by thispersondoesnotexist.com and similar sites to produce fake faces)

cc: @ZellaQuixote

@DrAdamAneevit

First up, we have @DrAdamAneevit, an antivax account with 38K followers and a GAN-generated face. Although this account's bio states that it is a "pArOdY", plenty of people are eating up its misleading tweets about COVID and vaccines.

@DrAdamAneevit

As it turns out, @DrAdamAneevit (permanent ID 949263131800424449) wasn't always called @DrAdamAneevit. Wayback machine archives reveal that this account was previously a "news" account named @/ForFactsSake101.

(archive:

https://web.archive.org/web/20190205193508/https://twitter.com/ForFactsSake101/status/1092869221095297026

)

@RowcliffeAndy

Next up is @RowcliffeAndy (permanent ID 1545105573997252608), an account created in July 2022 with a GAN-generated face. @RowcliffeAndy's greatest achievement thus far is spamming a meme until Dilbert creator @ScottAdamsSays amplified it.

@ViktorKoval9

Third on today's list of dodgy GAN-faced Twitter accounts is @ViktorKoval9 (ID 1497503918695530497). This account tweeted an image of a nonexistent tweet from a nonexistent blue-check account. The fake tweet contains a GAN-generated face, falsely represented as a boy in Ukraine.

@StevieDougie

Last, and possibly least, we have @StevieDougie (ID 324652265), a right-wing Twitter account that disapproves of Marxism and communism and has therefore (allegedly) moved from the United States to Vietnam, a country with a communist government.

https://twitter.com/conspirator0/status/1551305208285757445

How can we tell these four images are GAN-generated? There are a few tells: nonsensical backgrounds, mismatched/missing ears, weird artifacts where the hair meets the skin or background, and identical placement of the major facial features (as seen here:

https://twitter.com/conspirator0/status/1551305208285757445

)

https://twitter.com/conspirator0/status/1322704400226394112

Some tips on identifying GAN-generated faces:

https://twitter.com/conspirator0/status/1322704400226394112

Additional threads about GAN-generated faces and their use on (mostly) Twitter:

https://twitter.com/conspirator0/status/1457051817699860483

Note: the four accounts in this thread are presented as distinct examples of accounts that have GAN-generated face pics combined with other suspicious attributes. We found no evidence that the four accounts are related or run by the same entity.

@DrAdamAneevit

The operator of the @DrAdamAneevit account was apparently not a fan of this thread.

@RowcliffeAndy

The @RowcliffeAndy account has left a few replies to this thread. The attempt at falsely equating the use of a fake human face with the use of a cartoon avatar that no one would mistake for human has become a somewhat common tactic of bad-faith actors.

• • •

Missing some Tweet in this thread? You can try to force a refresh

This Thread may be Removed Anytime!

Twitter may remove this content at anytime! Save it as PDF for later use!

More from @conspirator0

Conspirador Norteño

@conspirator0

Mar 8

Just for fun, I decided to search Amazon for books about cryptocurrency a couple days ago. The first result that popped up was a sponsored listing for a book series by an "author" with a GAN-generated face, "Scott Jenkins".

cc: @ZellaQuixote

Alleged author "Scott Jenkins" is allegedly published by publishing company Tigress Publishing, which also publishes two other authors with GAN-generated faces, "Morgan Reid" and "Susan Jeffries". (A fourth author uses a photo of unknown origin.)

As is the case with all unmodified StyleGAN-generated faces, the facial feature positioning is extremely consistent between the three alleged author images. This becomes obvious when the images are blended together.

Read 5 tweets

Conspirador Norteño

@conspirator0

Feb 8

The people in these Facebook posts have been carving intricate wooden sculptures and baking massive loaves of bread shaped like bunnies, but nobody appreciates their work. That's not surprising, since both the "people" and their "work" are AI-generated images.

cc: @ZellaQuixote

In the last several days, Facebook's algorithm has served me posts of this sort from 18 different accounts that recycle many of the same AI-generated images. Six of these accounts have been renamed at least once.

The AI-generated images posted by these accounts include the aforementioned sculptures, sad birthdays, soldiers holding up cardboard signs with spelling errors, and farm scenes.

The common element: some sort of emotional appeal to real humans viewing the content.

Read 7 tweets

Conspirador Norteño

@conspirator0

Jan 24

As Bluesky approaches 30 million users, people who run spam-for-hire operations are taking note. Here's a look at a network of fake Bluesky accounts associated with a spam operation that provides fake followers for multiple platforms.

cc: @ZellaQuixote

This fake follower network consists of 8070 Bluesky accounts created between Nov 30 and Dec 30, 2024. None has posted, although some have reposted here and there. Almost all of their biographies are in Portuguese, with the exception of a few whose biographies only contain emoji.

The accounts in this fake follower network use a variety of repeated or otherwise formulaic biographies, some of which are repeated dozens or hundred of times. Some of the biographies begin with unnecessary leading commas, and a few consist entirely of punctuation.

Read 8 tweets

Conspirador Norteño

@conspirator0

Jan 10

It's presently unclear why, but over the past year someone has created a network of fake Facebook accounts pretending to be employees of the Los Angeles Dodgers. Many of the accounts in this network have GAN-generated faces.

cc: @ZellaQuixote

This network consists of (at least) 80 Facebook accounts, 48 of which use StyleGAN-generated faces as profile images. The remaining 32 all use the same image, a real photograph of a random person sitting in an office.

As is the case with all unmodified StyleGAN-generated faces, the main facial features (especially the eyes) are in the same position on all 48 AI-generated faces used by the network. This anomaly becomes obvious when the faces are blended together.

Read 6 tweets

Conspirador Norteño

@conspirator0

Oct 18, 2024

None of these chefs exist, as they're all AI-generated images. This hasn't stopped them from racking up lots of engagement on Facebook by posting AI-generated images of food (and occasional thoughts and prayers), however.

cc: @ZellaQuixote

These "chefs" are part of a network of 18 Facebook pages with names like "Cook Fastly" and "Emily Recipes" that continually post AI-generated images of food. While many of these pages claim to be US-based, they are have admins in Morocco per Facebook's Page Transparency feature.

Between them, these 18 Facebook "chef" pages have posted AI-generated images of food at least 36,000 times in the last five months. Not all of the images are unique; many have been posted repeatedly, sometimes by more than one of the alleged chefs.

Read 7 tweets

Conspirador Norteño

@conspirator0

Sep 8, 2024

Can simple text generation bots keep sophisticated LLM chatbots like ChatGPT engaged indefinitely? The answer is yes, which has some potentially interesting implications for distinguishing between conversational chatbots and humans.

cc: @ZellaQuixote
conspirator0.substack.com/p/baiting-the-…

For this experiment, four simple chatbots were created:

• a bot that asks the same question over and over
• a bot that replies with random fragments of a work of fiction
• a bot that asks randomly generated questions
• a bot that repeatedly asks "what do you mean by <X>?"

The output of these chatbots was used as input to an LLM chatbot based on the 8B version of the Llama 3.1 model. Three of the four bots were successful at engaging the LLM chatbot in a 1000-message exchange; the only one that failed was the repetitive question bot.