If you've been reading the analysis threads @ZellaQuixote and I post, you've probably seen the tweet schedule plots we include. Here's a simple web application you can use to make those plots yourself for accounts of your own choosing. makeadverbsgreatagain.org/allegedly
@ZellaQuixote The plots include 6-7 weeks of history ending with the most recent tweet, limited to the most recent 3200 tweets. If 3200 tweets don't go back the full 6-7 weeks, part of the plot will be blank. The circles are color-coded based on the app used to post the tweets.
@ZellaQuixote Update: a shaded "no data available" box will now be shown if the tweets available via the API (generally the most recent ~3200) do not go back a full 6 weeks. The goal is to prevent folks from mistaking the time range prior to the available tweets for a gap in activity.
@ZellaQuixote The percentage of tweets posted with each app or service has been added to the legend. Additionally, the overall percentage of tweets posted via automation apps/services is shown, along with a red check mark if over 90% (the threshold we've been using for "bot").
@ZellaQuixote Update: we added a test for 24/7 activity to makeadverbsgreatagain.org/allegedly. The score is the percentage of the displayed time range where the account tweets at least 24 hours straight with no gaps of 2 hours or longer. Accounts with scores of 50+% are flagged as automated (red check).
@ZellaQuixote Second update: a "Color Scheme" selector has been added. Two options:
Tweet source: tweets are colored by app/service used (this is the original behavior)
Tweet type: tweets are colored based on one of five categories (standalone tweet, retweet, quote tweet, reply, or thread)
@ZellaQuixote As we evolve this tool further, it's worth considering how to adjust the visualization parameters to make it harder for bad actors to use it to spread disinformation. These plots don't actually show suspicious activity, but the default options are being used to spin it that way.
@ZellaQuixote We've modified the site to automatically determine the appropriate scale for best visualizing the tweet volume of the account being analyzed (you can still manually select a different option if desired). Many thanks to @BeegorBucleor for inspiring this enhancement.
@ZellaQuixote@BeegorBucleor Update the fourth (we think?): we've added detection of repeated tweets to the site. A repetitiveness (is that a word?) score has been added to the legend, and a table of the tweets the account has most frequently repeated within their last 3200 is shown beneath the plot.
@ZellaQuixote@BeegorBucleor The repetition detection only applies to text content of tweets - links, images, video, etc are omitted, as are retweets. The repetitiveness score is calculated as follows:
@ZellaQuixote@BeegorBucleor Due to poor planning, we somehow failed to include @SeanSpammity in the example screenshots of the new "repeated tweets" feature. This has now been rectified.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
It's a day that end in "y", and a bunch of accounts with GAN-generated face pics are having "conversations" with each other about a game called "TheUnfettered" that apparently involves both NFTs and the Metaverse. #FridayShenaniGANs
These accounts are part of a network of 23 accounts with GAN-generated profile pics that (at least for the last year) tweet exclusively via the Twitter Web App. Most were created August 2020 or later, but six were made in 2011 and have old tweets sent via a variety of apps.
Unmodified GAN-generated face pics (at least so far) have the telltale trait that the major facial features are in the same position on every image. This becomes evident when we blend the profile images of the 23 accounts in the network.
#CanadaHasFallen is trending, and one of the more popular tweets is from an account with a GAN-generated profile pic: @youspecialagent, permanent ID 1277275248400969731.
Despite having only ~4500 followers, @youspecialagent's tweets featuring various hashtags opposing "vaccine passports" have repeatedly gone viral.
Over 1000 of @youspecialagent's followers appear to have been gained by posting a single followback tweet containing an anti-"vaccine passport" hashtag. A bunch of accounts replied to the tweet with the hashtag in question, spreading the hashtag further.
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.
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: