New OSoMe tools: Fall 2026 edition

October 28, 2026
New OSoMe tools: Fall 2026 edition

The Observatory on Social Media (OSoMe, pronounced “awesome”) has released several updates to its public tools for studying social media and online information: Botometer Blue, support for Threads in OSoMeNet, and a redesigned version of Fakey. In addition, we are announcing a new open-source social media simulator tool, SimSoM NGen.

Introducing Botometer Blue

Botometer has a new flavor: Botometer Blue. The new tool brings Botometer to Bluesky, calculating bot scores based on user profile data with a model similar to the one used in the past on Twitter. In addition, we are introducing a new “slop score” that estimates how closely an account’s recent posts resemble content generated by AI-powered social bots. Both scores use data retrieved through the Bluesky API. Like other Botometer scores, they are intended to identify signals of automation rather than provide a definitive classification.

The algorithm behind the slop score builds on our latest research for detecting AI-generated social bot content. In the paper, “Adversarial Creation and Detection of AI-Generated Social Bot Content,” to appear in the Proceedings of the ACM Conference on Information and Knowledge Management (CIKM), we developed an adversarial approach to mimic how bad actors could use large language models to impersonate real social media users. The resulting malicious digital twins were then used to generate a multilingual, cross-platform dataset of human and AI-generated messages. We show that models trained on this data outperform existing approaches for content-based bot detection on real-world data.

Threads comes to OSoMeNet

OSoMeNet is our network visualization tool for exploring patterns of information sharing and spread across social media platforms. It has been expanded to support Threads, alongside Bluesky, Mastodon, TikTok, and OSoMe’s historical archive.

OSoMeNet turns social media interactions into visual networks that help users examine how conversations take shape online. Due to current limitations of the API, the networks visualizing information diffusion on Threads tend to be sparse and do not include reposts or replies. Despite this, adding Threads gives researchers and other users another platform to study how information spreads, which accounts and communities interact, and where conversations gain traction. It also expands the range of social media environments that can be explored through a common visualization tool, making it easier to examine patterns across different platforms.

A redesigned Fakey

Fakey, a game that teaches media literacy while studying how people interact with information from low-credibility sources, has been redesigned to improve usability and more closely resemble a modern social media feed. We also added a Bluesky sharing option to the game summary and updated the tutorial with new screenshots and examples. An updated pipeline now pulls questionable news articles directly from the RSS/Atom feeds of audited publishers, providing a more reliable source of content for the game.

SimSoM NGen makes its debut

We’re excited to introduce SimSoM NGen, an agent-based model of social media platforms and users for conducting large-scale simulations of online information dynamics. What began as an experimental model is now an accessible tool for the broader research community. The open-source software is developed by the OSoMe team with fellow researchers at the University of Applied Sciences and Arts of Southern Switzerland and the University of Zurich.

Compared to previous simulators, SimSoM NGen is designed from the ground up to scale to large populations, include many ingredients of modern social media platforms, and be customized for exploring different research questions. The level of abstraction of SimSoM NGen strives to achieve a balance between high ecological validity and minimal complexity, enabling the causal attribution of observed phenomena. The software offers scholars a safe way to test how platform mechanisms, network structure, recommendation algorithms, user biases, coordinated inauthentic behavior, and moderation policies shape what people see and share. By making it easy to run “what if?” experiments on synthetic networks instead of real users, SimSoM NGen will help researchers better understand the vulnerabilities of the online public square, and how it might be made more resilient to manipulation.

The first version is geared for technical users, but stay tuned for updates, including a friendly web-based interface and a computational infrastructure to run simulations in the cloud.