How NewsGuard’s Misinformation Fingerprints Alert Companies and Governments to the Threats from Falsehoods Spreading Online
NewsGuard’s constantly updated catalog of machine-readable false narratives helps teams in the AI, digital-platform, national defense, and reputation-management sectors stay abreast of emerging falsehoods circulating online to track and mitigate the spread of misinformation.
Read on to learn more about the use cases of our Misinformation Fingerprints, and how they can help your team.
By Veena McCoole | Published on May 15, 2023
What are Misinformation Fingerprints?
Purpose-built for use by both human analysts and AI tools, NewsGuard’s Misinformation Fingerprints provide continuously updated alerts and analysis of the digital-information environment—and a powerful way to track narratives as they emerge and spread online.
Updated daily, this is the largest, most comprehensive catalog ever created of transparently sourced, provably false significant narratives circulating online. Microsoft trusts the fingerprints as ethically reported, responsible, and unbiased data to train Bing Chat’s large-language models (LLMs) to provide accurate responses to prompts on topics in the news and reduce their propensity to spread false claims. The Pentagon’s analysts at Cyber Command have been alerted to the latest Russian, Chinese, Iranian, and other hostile-information operations. Trust and safety teams at social media platforms are alerted to false narratives as they spread on any platform, giving them a needed tool to avoid their platforms being weaponized to spread harmful false narratives.
NewsGuard analysts extract and catalog the top misinformation narratives spreading online, adding dozens of new entries each week across topics including health, politics, election integrity, geopolitics, and war.
Each of the more than 1,300 “fingerprints” contains:
- A description of the false narrative
- Any variations of the false narrative
- A detailed debunk of the false narrative citing authoritative sources
- Information about the provenance of the false narrative and where and how it has spread (hence the “fingerprint”)
- Keywords and hashtags related to the narrative
- Boolean search terms designed to find new examples of the narrative while avoiding false positives
- A list of examples of content across multiple languages conveying the false claim, with excerpts and text snippets capturing precisely how the claim was articulated
- A categorization of the level of risk of harm from each narrative
- Filtering tools based on false-narrative topics and subjects
How can Misinformation Fingerprints be used?
- As data seeds for AI and social-listening tools, to trace false claims across the internet and social media. This strategy combines the precision and verifiability of human-sourced false narratives with the automated scale and reach of social-listening tools to create a scalable solution for ongoing threat monitoring and horizon scanning. To get started, explore NewsGuard’s Misinformation Risk Audit for generative AI companies.
- As a warning system providing alerts of new false narratives being promoted by state propaganda operatives and other malign actors for security and defense experts. NewsGuard partners at government agencies, security, and intelligence firms are alerted to new hostile-information operations spreading Misinformation Fingerprints—alongside our weekly Disinformation Risk briefings—for a comprehensive overview of the global state of disinformation, from state-media threats to emerging geopolitical falsehoods.
- By human analysts and content moderators to understand mis- and disinformation risks, and keep content moderation and threat-detection research teams informed about the latest false narratives circulating on the internet. An example of this is NewsGuard partner GIPHY, whose content-moderation teams regularly consult new false narrative entries to help them identify and mitigate emerging threats on GIPHY’s platform that could jeopardize users. Read more about GIPHY’s use of NewsGuard’s data in this case study.
- To empower end users of tech platforms, and facilitate the sharing of verified, accurate information among social networks. For example, the Ethos Network, an ethical social media app targeting Gen-Z users, has integrated NewsGuard’s Misinformation Fingerprints and associated “debunks” into its user interface, enabling users to verify the content of their posts before publishing them. By integrating NewsGuard’s data into the front-facing user experience, Ethos demonstrates its values of integrity, trust, and user safety to everyone on the platform.
