Conspirador Norteño

Conspirador Norteño is a Substack focused on analyzing and exposing the tactics behind social media manipulation, the spread of disinformation and misinformation, and the use of inauthentic accounts and AI-generated content. It provides insights into identifying fake social media activity and the implications of such activities on public discourse and elections.

Social Media Manipulation Disinformation and Misinformation Artificial Intelligence in Social Media Online Security and Privacy Social Media Platforms' Policies Analytical Tools and Techniques Elections and Political Campaigns

The hottest Substack posts of Conspirador Norteño

And their main takeaways

The red-to-blue disinformation pipeline

36 implied HN points • 28 Feb 25
🇺🇸 U.S. Politics Disinformation Election Media Political Campaigns Social media
  1. False claims were spread about Twitter users being part of a pro-Ron DeSantis group, but there was no real evidence to back this up.
  2. These misleading stories originated from a blog by a Republican, and later Democrats used them to accuse other users without proof.
  3. The misinformation keeps spreading, showing how easy it is for false narratives to be believed and shared, especially on social media.

These politically enthusiastic supplement fans do not exist

44 implied HN points • 22 Feb 25
🇺🇸 U.S. Politics Social media Disinformation Health Consumer Rights Advertising
  1. Some social media accounts are fake and use AI-generated faces to push products, like weight loss supplements. They create the impression of being real people to promote their ads.
  2. These accounts often mimic authentic political personas, but they show signs of being inauthentic, like having repeated text or the same profile images. This raises questions about the trustworthiness of the content shared.
  3. A website linked to these ads uses fake reviews from AI-generated profiles, making it seem like people are satisfied customers when they're not real. This highlights the need to be careful about schemes like these online.

Spam in the firehose, part III

32 implied HN points • 14 Feb 25
🕹 Technology Social media Spam Detection User Behavior Online Trends
  1. There is a large network of fake accounts on Bluesky, all created to follow real users quickly and in groups. These accounts look almost identical, using default profile images and very little personal information.
  2. The spam accounts often follow a mix of different types of real accounts, including celebrities and businesses, but they usually don't have much content themselves. They mainly repost or reply with very simple comments.
  3. The way these fake accounts behave, like following in large swarms, makes it easy to spot them. They create a clear pattern that shows they aren't real users.

Antisocial media

48 implied HN points • 08 Feb 25
🕹 Technology Social media AI Ethics Digital Culture Content creation
  1. Many Facebook accounts post AI-generated images that trick users into feeling emotions like sadness or sympathy. These images often look real but are just made by computer programs.
  2. The same AI images get shared by different accounts, leading to repetitive and spammy content on the platform. Users might see the same sad story or image posted multiple times.
  3. Some of these accounts create stories to go with their images, making them seem more genuine. But it's all part of an effort to capture attention using artificial content.

Selling followers where the skies are blue

56 implied HN points • 24 Jan 25
🕹 Technology Social media Digital marketing Online Security Internet Trends Fake Accounts
  1. There are fake follower selling services popping up on social media platforms like Bluesky. These services exploit the platform's growth by creating fake accounts to sell followers, likes, and shares.
  2. A large number of fake accounts have been created quickly and they often share similar biographies, usually in Portuguese. This suggests they are not real users, but rather automated bots.
  3. Buying followers is not a good idea. Most accounts gained through these services have little to no genuine followers and their involvement can harm your social media presence.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Finding fake followers (Bluesky edition)

68 implied HN points • 18 Jan 25
🕹 Technology Social media Internet Data Analysis Spam Detection Artificial Intelligence
  1. You can spot fake followers on Bluesky by looking for accounts with similar join dates and generic profiles. These accounts often have no posts and repetitive bios.
  2. Using a method where you track the followers of suspected fake accounts can help identify whole networks of fake followers. By downloading and filtering their followers, you can map out these networks.
  3. The Bluesky platform has a real-time feature called the firehose that makes it easier to catch fake follower activity as it happens. However, this can give some false positives, so users need to be careful.

These defenders of democracy do not exist

76 implied HN points • 05 Jan 25
🇺🇸 U.S. Politics Social media Democracy AI Technology Political Activism
  1. Some social media accounts pretend to be liberal activists but are actually fake and use AI to generate responses. They try to look real but their activity is suspicious.
  2. These accounts follow a lot of people quickly but don't gain many followers back. This pattern shows they are likely bots trying to manipulate engagement.
  3. Many of the replies from these accounts are generic offers of help or unsolicited responses. It's often clear they are automated replies, not real conversations.

These Dodgers employees do not exist

60 implied HN points • 10 Jan 25
🕹 Technology Social media AI Online Privacy Digital Culture Fake news
  1. There are fake Facebook accounts that pretend to be Los Angeles Dodgers employees. They use AI-generated faces and photos of real people.
  2. These fake accounts haven't posted much content and mostly just have profile pictures. Many of them are friends with each other online.
  3. The purpose of these accounts is unclear, but they often check into random locations, which may not mean anything. It's a strange situation.

Spam in the firehose

128 implied HN points • 06 Dec 24
🕹 Technology Social media Spam Detection Data Analysis Online Security Programming
  1. Monitoring the Bluesky firehose can help quickly spot fake accounts. By looking for repeated names and profiles, it's easier to identify spam activity.
  2. A large number of spam accounts often share similar biographies. One group had over a thousand accounts with variations of the same few phrases.
  3. Many spam accounts use stolen images as profile pictures. This makes them look less authentic and easier to identify as spam.

The $94,900 X account

12 implied HN points • 31 Jan 25
🕹 Technology Social media Digital Assets Online Marketing Cryptocurrency
  1. Buying a social media account like @crypt0mutant for $94,900 is a risky move. There's a good chance it won't work out and the account might get banned.
  2. These accounts have a history of being renamed and repurposed, which makes it hard to trust them. Many followers might not even be real.
  3. The accounts post repetitive replies and promote random crypto coins, but they don't get much engagement. This suggests their followers aren't very active or interested.

“A title similar to the curious content that makes users click to see”

48 implied HN points • 15 Dec 24
🕹 Technology Social media Spam Detection AI Impact Content creation Plagiarism
  1. Social media networks are being flooded with spammy posts that link to fake news sites. These are typically run by accounts that pretend to be real people using stolen profile images.
  2. Many spam accounts were created in batches, suggesting automated processes are involved. This makes it easier to spot the fake accounts since they often post the same content repeatedly.
  3. The spam accounts are not limited to one platform; they are also found on different social media like X (formerly Twitter). They mainly promote articles that are simply copied from popular news sites.

Bogus accounts in the blue skies

80 implied HN points • 16 Nov 24
🕹 Technology Social media Digital Culture Online Privacy User Behavior
  1. Bluesky is getting a lot of new users, but some of those accounts are fake. These inauthentic accounts often use the same photos and bios, making them easy to spot.
  2. The bogus accounts post similar things, like photos of fitness centers and coffee, and use identical captions. This makes it clear they aren't real people.
  3. These fake accounts seem to follow popular liberal users from X and try to blend in. As more people join Bluesky, the influence of these inauthentic accounts could grow.

This landscape does not exist

32 implied HN points • 28 Dec 24
🕹 Technology AI Social media Digital Content Spam Detection Online behavior
  1. Facebook is flooded with AI-generated images, often coming from pages that aren't run by regular users. These images sometimes get a lot of attention, but they mainly come from content farms.
  2. Many Facebook pages post the same AI-generated images around the same time, with slight changes to avoid detection. This suggests they might all be operated by the same group of people.
  3. The AI-generated images often look strange and unrealistic, with obvious glitches like odd-looking roads and animals. They tend to have brighter colors than real photos, making them easy to spot.

For You

44 implied HN points • 22 Nov 24
🕹 Technology Social media Algorithms Data Analysis User Experience Content curation
  1. The 'For You' feed on X shows mostly posts from accounts you don't follow. In fact, more than half of the recommended posts come from these unfamiliar sources.
  2. Elon Musk's posts are the most frequently suggested, even to users who do not follow him. This indicates that trending figures often dominate the recommendation algorithm.
  3. Connections between suggested accounts are mostly based on repost interactions. Most recommended accounts have links to the ones you already follow, showing a network effect.

The curious case of the Election Day follower exodus

52 implied HN points • 08 Nov 24
🇺🇸 U.S. Politics Elections Social media Political Behavior Media influence
  1. Many people lost followers on X after the recent election, and it's happening to mostly liberal accounts. Some of these losses seem to come from real users quitting.
  2. In contrast, prominent conservative accounts mostly gained followers during this time, indicating a shift in engagement on the platform.
  3. There was also a crackdown on fake accounts leading to follower loss, which means some losses are due to accounts being banned rather than people leaving.

Selling account names in the blue, blue skies

36 implied HN points • 28 Nov 24
🕹 Technology Social media Online Scams Data Analysis Internet culture
  1. Handle squatting is when people register social media handles to sell them later. Even though Bluesky allows custom domain names as handles, some still try to squat.
  2. Buying account names is risky and usually a bad idea. It's better to create your own accounts instead of getting them from spammers.
  3. Some recent accounts on Bluesky show repetitive bios and were created in batches, indicating possible spam activity. One such account even changed its bio to seem more legitimate.

Spam in the firehose, part II

20 implied HN points • 21 Dec 24
🕹 Technology Social media Internet Security Online behavior Digital marketing Spam Detection
  1. A large network of spam accounts has been created on the platform Bluesky, with over 15,000 accounts active. These accounts often have similar or identical profile photos and biographies.
  2. The most common biographies used by these spam accounts are repetitive and lack originality. Many accounts use phrases like 'passionate about exploring innovative ideas' interchangeably.
  3. These spam accounts mostly repost content from varied accounts but don’t create original posts themselves. There is also a suspicious account promoting social media services that seems linked to this spam network.

Community Notes and potential vote brigading

36 implied HN points • 02 Nov 24
🇺🇸 U.S. Politics Voting Systems Data Analysis Community Engagement
  1. Community Notes on the X platform use a unique voting system to check facts, requiring a mix of helpful ratings. This makes it harder to manipulate which information is shown.
  2. Recent voting patterns show large bursts of upvotes or downvotes after political posts, often favoring right-leaning perspectives. This suggests some users might be trying to game the system.
  3. Out of many notes reviewed, most aimed to correct or add context to political content. While some notes were rated 'helpful,' others still need more varied ratings to be visible.

This barndominium does not exist

44 implied HN points • 01 Mar 24
🕹 Technology Social media AI Online Security Ethics
  1. The Facebook account 'Barndominium Gallery' is posting AI-generated images of houses to engage users, many of whom believe the images are real
  2. The AI-generated images of 'barndominiums' contain synthetic artifacts, like unrealistic fire placements and impossible elements, that reveal their inauthenticity
  3. The 'Barndominium Gallery' account operator solicits personal information from users in the comments, risking exposure to potential privacy threats or fraud

Multilevel marketing mayhem

44 implied HN points • 25 Feb 24
💼 Business Marketing Online business Digital marketing Entrepreneurship
  1. Be cautious of 'online businesses' that prompt you to set up websites with identical language and testimonials, often part of multilevel marketing schemes.
  2. Identical sales pitches on multiple websites can be easily found through internet searches, revealing potential interconnectedness.
  3. Participating in these schemes may lead to continuous spam emails, financial losses, and the risk of personal information being compromised.

Love, happiness, blessings, and spam

32 implied HN points • 16 Mar 24
🕹 Technology Social media Artificial Intelligence Data Analysis Content creation Social media Data Analysis Fake Accounts
  1. Spam accounts use repetitive and fake positive messages to amplify content, making it appear more popular than it actually is.
  2. Researchers are now facing difficulties in mapping out spam account networks due to limitations in data access.
  3. Spam network accounts use GAN-generated faces and peculiar vowels in account names, creating an association with suspended spam networks.

Seeing isn't always believing: video edition

36 implied HN points • 17 Feb 24
🕹 Technology AI Generative AI Deepfakes
  1. The advancement of text-to-video generative AI like Sora raises concerns about deceptive video content, introducing the concept of the "liar's dividend."
  2. Despite impressive quality, AI-generated videos by Sora exhibit anomalies that reveal their synthetic origins, such as sudden appearance and disappearance of objects.
  3. While AI-generated videos can be photorealistic, they often contain telltale signs of synthetic generation, cautioning against an excessive distrust of all videos and emphasizing the long-standing history of manipulating video content.

GAN-tastic politics enthusiasts

40 implied HN points • 06 Jan 24
🕹 Technology AI Social media Spam Network
  1. A network of fake accounts with GAN-generated faces is spamming users with phrases like 'politics enthusiast'.
  2. The fake accounts have repetitive behaviors and characteristics, such as long handles with many vowels and repetitive use of certain phrases.
  3. Some replies from the fake accounts show signs of being artificially generated, with error messages and tagging issues.

Happy Holidays

40 implied HN points • 24 Dec 23
🇺🇸 U.S. Politics Misinformation Social media
  1. The blogger will continue exploring deceptive uses of generative AI related to the 2024 U.S. election season.
  2. There will be a focus on dissecting and debunking misleading coverage of misinformation in mainstream publications.
  3. Less content will be dedicated to consolidating older threads from Twitter, with a shift towards discussing other platforms like Bluesky and Mastodon.

The curious case of user one134500 and the most obvious reply spam ever

68 implied HN points • 22 May 23
🕹 Technology Social media Cryptocurrency Spam Fake Accounts Online Security
  1. A network of spam accounts flooded Twitter with repetitive replies containing login credentials for a suspicious website.
  2. Over 13,000 active accounts in this spam network were created in March or April 2023 and only post replies.
  3. Spam campaign targeted various accounts including Reuters, Elon Musk, and news organizations, with a high percentage of replies in English.

GET WOKE UP by these nonexistent authors

36 implied HN points • 26 Nov 23
📰 News Satire
  1. An "anti-woke" satirical news site used AI-generated faces for its non-existent authors.
  2. The GAN-generated faces on the site originated from generated.photos with watermarks removed.
  3. Social media accounts linked to the site also use the same GAN-generated faces as the authors.

Trending topics and word salad

20 implied HN points • 10 Mar 24
🕹 Technology Social media Cryptocurrency Spam Algorithm Automation
  1. Trending topics on social media can be manipulated by spam posts containing random words instead of coherent sentences.
  2. Accounts participating in spam trends can show signs of being hijacked and may switch focus from personal topics to spam suddenly.
  3. Past spam campaigns involving hashtags and random word mashups have been successful in manipulating social media trends.

Some thoughts on social media, disinformation, and the 2024 elections

32 implied HN points • 12 Nov 23
🇺🇸 U.S. Politics Social media Disinformation Elections Research Media Coverage
  1. Data-driven research on social media activity for the 2024 elections is made difficult due to factors like reduced transparency by platforms and attacks on researchers.
  2. Generative AI advancements have made it easier to create misleading content like fake images and videos.
  3. Media coverage of disinformation and social media manipulation may not always be based on legitimate research, highlighting the importance of skepticism.

Anatomy of an online outrage mob

72 implied HN points • 13 Feb 23
🇺🇸 U.S. Politics Social media Online behavior Twitter Harassment Threats
  1. Right-wing social media influencers can cause online outrage mobs by tagging institutions on Twitter.
  2. 72 school district, university, and hospital accounts experienced a significant spike in mentions after being tagged by an influencer.
  3. The barrage of tweets following these tags often include threats of real-world violence, violating Twitter's terms of service.

Flashback Friday: Taxi!

20 implied HN points • 09 Feb 24
🕹 Technology Social media Automation Data Analysis
  1. A network of taxi and real estate-themed social media accounts were used to boost political content on Twitter through automation.
  2. The botnet consisted of at least 98 Twitter accounts with automated posting schedules that operated 24/7.
  3. The botnet retweeted content based on hashtags, focusing on small accounts and political tweets rather than popular ones.