CRED-1: Building an Open Domain Credibility Dataset for Misinformation Pre-Bunking

2026-03-21 · Source: Machine Learning on Medium · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Software Development & Engineering · Depth: Intermediate, short

Summary

CRED-1 is an open, multi-signal credibility dataset designed for pre-bunking online misinformation, developed as part of doctoral research at Frankfurt University of Applied Sciences. Unlike traditional binary credibility lists, CRED-1 assigns a composite score from 0.0 to 1.0 to 2,672 domains by integrating signals such as source labels from OpenSources.co and Iffy.news, domain age, web popularity via Tranco Top-1M, Google Fact Check frequency, and Google Safe Browsing threat intelligence. This approach addresses the limitations of static, opaque, single-label systems by providing nuanced, verifiable, and configurable credibility assessments. The dataset is compact, enabling on-device, privacy-preserving lookups for applications like browser extensions, and is fully reproducible using Python scripts.

Key takeaway

For AI Scientists and Research Scientists developing tools to combat online misinformation, CRED-1 offers a robust, open-source dataset for pre-bunking. You should consider integrating its multi-signal scoring approach into your models or applications to provide more nuanced and privacy-preserving credibility assessments, moving beyond simplistic binary labels. The dataset's reproducibility and compact format facilitate easy adoption and customization for various research and deployment scenarios.

Key insights

Multi-signal credibility scoring offers a nuanced, dynamic, and transparent alternative to binary domain labeling for misinformation pre-bunking.

Principles

• Credibility is multi-faceted, not binary.
• Pre-bunking is more effective than debunking.
• On-device processing enhances privacy.

Method

CRED-1 computes a composite credibility score by weighting and combining independently verifiable signals like domain age, web popularity, fact-check frequency, and threat intelligence, allowing for configurable thresholds.

In practice

• Embed compact JSON in mobile apps.
• Integrate into browser extensions for warnings.
• Use for content moderation pipelines.

Topics

• CRED-1 Dataset
• Multi-Signal Scoring
• Misinformation Pre-Bunking
• Domain Credibility
• Online Misinformation

Code references

• aloth/trackless-links
• aloth/cred-1

Best for: AI Scientist, Research Scientist, AI Researcher, Data Scientist, Software Engineer

Related on AIssential

• SynCred-Bench: Benchmarking Synthetic Credibility in AI-Generated Visual Misinformation — AI-generated visual misinformation with synthetic credibility poses a severe, underexplored threat, as current detection systems and humans struggle to identify it reliably.
• Phantoms and Disclosures: a Causal Framework for Auditing Synthetic Data — The framework distinguishes true from phantom data disclosures in synthetic data using statistical testing, requiring no model access.
• Dissecting Subjectivity and the "Ground Truth" Illusion in Data Annotation — Human disagreement in data annotation is a vital sociotechnical signal, not mere noise.
• RECOM: A Validity Discrimination Tradeoff in Automatic Metrics for Open Ended Reddit Question Answering — Automatic metrics for open-ended QA exhibit a validity–discrimination tradeoff, failing to both confirm content alignment and reliably rank systems.
• TraceBack: Multi-Agent Decomposition for Fine-Grained Table Attribution — TraceBack provides fine-grained, cell-level attribution for table QA using a multi-agent decomposition framework.
• Probing Multimodal Large Language Models on Cognitive Biases in Chinese Short-Video Misinformation — MLLMs struggle with short-video misinformation due to cognitive biases like authority and popularity effects, despite multimodal reasoning capabilities.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine Learning on Medium.