Every quarter, we package the structured datasets generated by the Viqus engine — thousands of AI-scored news records, entity graphs, and verdict breakdowns, ready for research, training, and analytics.
Each quarterly release is a single, self-contained viqus_export.jsonl file with nested analysis. Unzip and query.
Every AI news story with its original title, source, publication date, sentiment classification, extracted entities, and keywords.
viqus_export.jsonlAI-generated deep analysis: refined headline, category, key points, detailed summary, and a "why it matters" contextual breakdown.
Analysis EngineDual-axis scoring separating Media Hype from Real Impact (1–10), with a verdict title and full AI analysis explanation.
Scoring LayerEvery quarter we package and publish a new drop. All past and current releases are available for free download from the archive.
.zip bundles with README, LICENSE, schema & data
CC-BY 4.0 · Free for any use with attribution
Every record follows a strict, documented schema. Ready for pandas, Spark, or any JSONL-compatible pipeline.
idintUnique record IDtitlestringOriginal headlinesource_namestringPublisher namedate_publishedISO8601Publication timesentimentenumPositive / Negative / Neutralentitiesjson[]{name, type} arraykeywordsjson[]Contextual tagsanalysis_jsonobjectFull AI analysis↳ headlinestringAI-refined title↳ categoryenum6 Viqus verticals↳ viqus_verdictobjecthype_score, impact_score, title, ai_analysis{ "id": 1, "title": "Anthropic will start training...", "link": "https://theverge.com/...", "source_name": "The Verge AI", "date_published": "2025-08-28T12:00:00", "sentiment": "Negative", "entities": "[{\"name\":\"Anthropic\",\"type\":\"Company\"}]", "keywords": "[\"AI\",\"Data Privacy\",\"Opt-Out\"]", "html_filename": "news/anthropic-shifts-to...", "analysis_json": { "headline": "Anthropic Shifts to User Data", "category": "Ethics & Society", "summary": "Anthropic is changing...", "key_points": ["...", "..."], "why_it_matters": "...", "detailed_summary": "...", "viqus_verdict": { "title": "Data Dependence: A Growing Risk", "hype_score": 6, "impact_score": 8, "ai_analysis": "While the shift..." }, "created_at": "2025-08-29T01:23:15" } }
From raw feed to packaged dataset — fully automated, locally processed, human-reviewed.
Continuous monitoring of 200+ technical feeds, repos, and channels.
Local LLM scoring via Ollama. Dual-prompt verdict, entities, key points.
Schema enforcement, deduplication, statistical quality checks.
JSONL export with README, LICENSE, schema docs, SHA256 hash.
Study AI industry trends, media coverage patterns, and the hype-vs-reality gap. Longitudinal analysis made effortless.
Use enriched JSONL records as training data for domain models or a structured knowledge base for RAG.
Feed drops into BI tools, Jupyter notebooks, or custom dashboards for category trends and scoring distributions.
Track competitor mentions, technology adoption signals, and market momentum across the AI ecosystem.
We release datasets quarterly. Subscribe to receive a signal the moment each new drop goes live.
We respect your privacy. Choose which cookies you allow us to use. You can change these settings at any time via the link in the footer. Essential cookies are required for the site to function.
Required for basic site functionality, security, and accessibility. Cannot be disabled.
Help us understand how visitors interact with our site by collecting anonymous usage data via Google Analytics.
Enable personalized advertising and remarketing features across Google services.
Allow personalized content and ad recommendations based on your browsing behavior.