Information Operations Exploit APIs to Manipulate Social Media

arXiv:2610.02591v1 Announce Type: new Abstract: Research on information operations has focused primarily on the coordinated accounts involved in such campaigns and in the content they promote. Bad actors manage the activity of inauthentic accounts through programmatic interfaces (APIs), but the role of this underlying infrastructure in enabling and uncovering such coordinated behaviors remains largely unexplored. This study investigates how the Twitter (now X) API was used to coordinate influence operations on the platform. Since API access is managed through developer apps, we extract app metadata from 43 operations taken down between 2018 and 2021 and apply three complementary computational approaches to study diverse forms of manipulation. Our analysis reveals several recurring patterns: API-as-a-service infrastructure reused across campaigns, gaming apps performing unauthorized engagement without user consent, spoofed app names to hide activity origins, and synchronized app networks indicating centralized control. Adding app usage as a coordination indicator provides a modest but consistent improvement in detection. These findings suggest that social media APIs play an important role in enabling online manipulation, and that the removal of app-level metadata by Twitter/X has eliminated an important tool for platform transparency and moderation.

APIs enable manipulation

This is historical forensic research on pre-2021 Twitter manipulation infrastructure, not a claim about current AI capability. It documents how platform API design choices affect detectability of coordinated inauthentic behavior, implying reduced transparency going forward.

The study directly evidences how information operations utilize APIs to conduct coordinated manipulation of social media, impacting public discourse. The removal of app-level metadata by platforms like X hinders transparency and moderation, exacerbating the problem. This represents a significant degradation of the information environment.

The study shows that APIs were used to manipulate social media, reducing transparency and making detection harder. This undermines trust in social media platforms.

SOURCE     arXiv cs.CY (Computers and Society) (tier 1)
   ↓
DOCUMENT   d366171b-7b73-4508-909e-e2e3d2cf7861
           https://arxiv.org/abs/2610.02591
   ↓
EVIDENCE   3 extracted excerpts
   ↓
MODEL RUN  4 runs, methodology 1.0.1
   ↓
SCORE      -65  (Strongly adverse)
   ↓
CONFIDENCE 80%