It's February, and a group of accounts with GAN-generated profile pics are tweeting identical "Merry Christmas" tweets promoting some kind of NFT giveaway.
(GAN = "generative adversarial network", the AI technique used by thispersondoesnotexist.com to produce fake faces)
These accounts are part of a spam network consisting of (at least) 6105 accounts created in February 2022, mostly in large batches. All have GAN-generated profile pics. The same names are reused across accounts - the 6105 accounts only have 25 unique first names between them.
The GAN-generated face pics produced by tools like thispersondoesnotexist.com have the telltale trait that the major facial features (particularly the eyes) are in the same position on every image. This becomes obvious when the images are blended together.
Despite the fact that one can generate as many unique GAN faces as one desired, this spam network frequently uses the same fake face on multiple accounts.
The accounts in this network have followed several accounts en masse, mostly metaverse and/or NFT-themed. Former Ukraine President Petro @Poroshenko, current US President Joe Biden (@POTUS), and Tesla founder @elonmusk also turn up.
The accounts in this network (allegedly) tweet exclusively via the Twitter Web App. Their tweet content is a mix of original tweets, retweets, and replies.
The "original" content tweeted by this network consists of two tweets that have been repeated verbatim by hundreds of the accounts. (One of these tweets is the February "Merry Christmas" NFT giveaway tweet mentioned earlier.)
The network's retweets are also repetitive - almost all retweets are of the same four NFT/crypto tweets. The network's replies are all brief threads containing the phrase "Fishing Free NFTs" and tagging @Fishing_Crypto and large numbers of seemingly random accounts.
More information on GAN-generated profile pics in this compliation of threads:
The 2022 #WinterOlympics are underway, and spammy networks such as this group of accounts with anime avatars tweeting about the "pokesperson of Chinese Embassy in the US" are taking an interest in the occasion.
These tweets were posted by a network consisting of (at least) 152 similarly-named accounts created in batches between September 2021 and January 2022. All tweet (allegedly) via either the Twitter Web App or Twitter for Android, and all have few or no followers or followees.
Many of the tweets posted by this network are duplicated on multiple accounts. The main theme is denial of human rights abuses against Uyghurs, with "feel-good" content about Xinjiang thrown in. The network frequently uses Olympic hashtags (#WinterOlympics, #Beijing2022, etc).
For a change of pace, I'm doing a bit of a YouTube experiment. I'm going to choose six songs of varied genre on a new YouTube account, and then just listen to whatever it creates as the "My Mix" playlist. #FruitOfTheAlgorithm
Unmodified GAN-generated face pics have the telltale trait that the major facial features (particularly the eyes) are in the same position on every image, and @Gabby_ucm's profile pic is no exception. There are also anomalies in the teeth, clothing, and hair of @Gabby_ucm's pic.
More on GAN-generated images and their use on Twitter in this set of threads:
What do 2022 Congressional candidates @BlakeHarbinGA (R), @DavidGiglioCA (R), and @RajiRab2020 (D) have in common? All three have recently been followed by thousands of newly-created accounts with lowercase display names and zero tweets.
The new followers of the three Congressional candidates are part of a botnet consisting of (at least) 38105 accounts with random-looking names created in December 2021/January 2022. None of the accounts has ever tweeted, though most have liked dozens or hundreds of tweets.
The accounts in this botnet have extremely repetitive biographies, using only 1085 unique biographies across 38105 accounts.
In one of the sillier developments in recent history, the adherents of this flawed "vetting guide" are now issuing "alerts" about my account because (gasp!) a small number of the 41K accounts following me look suspicious.
Since I'm now being targeted, I'm going to be blunt. The principal author of this "vetting guide", @EnseySherwood, blatantly makes shit about about bot detection, such as this false claim that servers have a setting that Twitter could use to eliminate all bots in 10 minutes.
If you have been relying on this person or the "vetting guide" for advice on bots and suspicious accounts, you've been fooled, and you should probably also give some thought as to why certain "vetting experts" constantly lock their accounts and refuse to engage with critics.
It's possible to automate this technique and download most or all of an account's tweets archived on Wayback Machine/Internet Archive. Here's how to write a basic Python program to download the tweets and store them as a CSV file.
Step 1: read the list of all the Wayback Machine archives of the account in question. The URL used is the same call the Wayback Machine website uses, but the maximum number of results has been bumped to 1 million.(Wayback Machine defaults to 10K max results.)
Step 2: filter the results to archives of individual tweets or replies (throwing away things like archives of someone's profile etc). Tweets from before Twitter switched to using Snowflake IDs (November 2011 blog.twitter.com/engineering/en…) are also discarded.