What is AVALW News?
AVALW News is a global news synthesis platform that uses artificial intelligence to produce original journalism from thousands of verified sources. For every story we publish, our system reads reporting from 10 to 20 or more independent outlets, extracts confirmed facts, cross-references them across multiple publications and languages, and generates a completely new article.
We are not an aggregator that collects headlines. We do not copy, translate, or paraphrase any single source. Every article on AVALW News is an original work, created by synthesizing verified facts into a new narrative with its own structure, phrasing, and editorial perspective rooted in factual accuracy.
Every source used in a synthesis is fully credited with a direct link to the original reporting. Our readers can always trace any fact back to its origin.
The numbers
How it works
Every article published on AVALW News passes through a five-stage pipeline before it reaches you.
Listen
Our indexing engine continuously monitors over 10,000 RSS feeds from news organizations in more than 50 countries. Sources span every major language and region, from national broadcasters to independent investigative outlets. We respect robots.txt directives and honor publisher opt-out requests at all times. The system refreshes every five minutes, ensuring coverage of breaking events within minutes of publication.
Cluster
When multiple sources publish stories about the same event, our clustering engine groups them together. A political development in Brussels might be reported by outlets in English, French, German, Romanian, and Spanish simultaneously. Our system recognizes these as the same story and merges them into a single cluster, regardless of language, ensuring the broadest possible view of the event.
Verify
From each cluster, the system extracts structured facts: who is involved, what happened, where it took place, when it occurred, why it matters, and how it unfolded. Every fact is cross-checked against the full cluster. A claim is only included in the final article if at least three independent sources confirm it. Unverified claims, single-source allegations, and unattributed speculation are excluded.
Write
Using a locally hosted large language model, the system generates a completely new article based on the verified facts. The result is original text with its own structure, its own phrasing, and its own narrative arc. No sentences are copied from any source. The output reads like professional journalism because it follows the same principles: lead with the most important facts, provide context, present multiple perspectives, and cite sources throughout.
Illustrate
Every image on AVALW News is generated by artificial intelligence running on our own hardware. We never use photographs from news sources. All AI-generated images are clearly labeled as illustrations and include embedded metadata identifying them as synthetic content, in compliance with the EU AI Act. Images are designed to be atmospheric and editorial in nature, never to depict specific individuals or simulate real photographs.
Our principles
Transparency
Every article lists every source used in its creation. Every image is marked as AI-generated. Every step of the editorial process is documented publicly in our Terms of Service. There is no hidden process.
Facts over opinion
We extract and verify facts. We do not add commentary, editorial judgment, ideological framing, or political perspective. When sources disagree, we present both sides and let the reader decide.
Full attribution
We link to every original article that contributed to a synthesis. We drive traffic back to the publishers who do the original reporting. We believe good journalism deserves to be found, read, and credited.
Privacy by design
We do not collect IP addresses, browser fingerprints, device information, or geolocation data. We use only strictly necessary cookies for basic site functionality. We do not run analytics or tracking of any kind.
Accountability
Every article includes a report button. Every report is reviewed by a human editor within 24 hours. If we get something wrong, we correct it publicly and promptly.
Independence
Our AI models run on our own servers. We do not rely on third-party AI providers or cloud APIs. Our editorial pipeline is entirely self-hosted, which means no external entity has influence over what we publish or how we publish it.
Copyright and legal compliance
AVALW News is built from the ground up to respect intellectual property law. We do not copy articles. We do not paraphrase articles. We do not republish content from any source. Here is what we do, and why it is legal.
What is protected by copyright, and what is not
Copyright law protects the specific way a journalist writes a story: the choice of words, the sentence structure, the narrative arc, the creative expression. Copyright does not protect the underlying facts. The Berne Convention, Article 2(8), states explicitly: "The protection of this Convention shall not apply to news of the day or to miscellaneous facts having the character of mere items of press information."
This means that the fact "the president rejected the peace proposal on Sunday" is not owned by anyone. The specific paragraph a journalist wrote to report that fact is protected. AVALW News uses only the facts. We never use the expression.
How we are different from illegal aggregators
| Criteria | Illegal copying | AVALW News |
|---|---|---|
| Source count | Takes from 1 article | Synthesizes from 10 to 20+ independent sources |
| Process | Copies or paraphrases text | Extracts only facts, writes entirely new text |
| Creative expression | Preserves or rewrites original phrasing | Generates completely new structure and phrasing |
| Images | Copies photos from sources | 100% AI-generated, clearly labeled |
| Attribution | Minimal or absent | Full attribution with direct links to every source |
| Effect on market | Replaces the original, takes traffic away | Drives traffic back to original publishers |
| Legal status | Copyright infringement | Transformative use, legally protected |
The legal framework we operate under
Our operations comply with multiple overlapping legal frameworks:
EU Copyright Directive (2019/790)
Article 15 protects press publications but explicitly exempts "individual words or very short extracts." We extract only factual data points, never creative expression. Our output is a new, transformative work.
Berne Convention (Art. 2.8)
The international treaty signed by 182 countries states that "news of the day" and "miscellaneous facts having the character of mere items of press information" are not protected by copyright.
EU AI Act (Art. 50)
We comply fully with AI transparency obligations. All AI-generated content is clearly labeled. All images carry machine-readable metadata identifying them as synthetic. Users can report errors at any time.
U.S. Fair Use (17 U.S.C. 107)
Under the four-factor fair use test, our synthesis is transformative in purpose, uses only factual (not creative) material, takes no copyrightable expression, and benefits the market for original journalism by driving traffic to sources.
GDPR (EU 2016/679)
We collect no personal data. No IP addresses. No browser information. No tracking. No analytics. Only strictly necessary cookies for basic site preferences.
DMCA Safe Harbor (17 U.S.C. 512)
We maintain a designated DMCA agent, a formal notice-and-takedown procedure, and a repeat infringer policy. Any publisher can request content removal and receive a response within 48 hours.
Publisher opt-out
Any publisher who does not wish to be included in our synthesis can opt out at any time through any of the following methods:
- Add a robots.txt rule:
User-agent: AVALWBot / Disallow: / - Add a meta tag:
<meta name="AVALWBot" content="noindex"> - Email office@avalw.com (Subject: Publishers) with the subject "Opt-out Request." We process all requests within 48 hours.
We believe in cooperation with publishers, not competition. Our platform exists to help readers discover stories they would otherwise never find, and to send those readers directly to the original reporting.
The technology
AVALW News runs on infrastructure wholly owned and operated by AVALW S.R.L. in the European Union. All AI processing happens on-device, not through external APIs. This architecture ensures editorial independence, data sovereignty, and compliance with EU regulations including the AI Act and GDPR.
Our core systems include a news indexing engine that processes over 10,000 RSS feeds on a five-minute cycle, a cross-language clustering system that detects the same event across dozens of publications, a fact extraction pipeline that structures raw reporting into verifiable claims, a locally hosted large language model for article synthesis, and a Stable Diffusion image generation system running on dedicated GPU hardware.
All infrastructure is containerized and runs across dedicated servers in Romania, within the European Union's regulatory framework.
About AVALW S.R.L.
AVALW S.R.L. is a technology company based in Iasi, Romania, founded in 2019. We build AI-powered platforms for information discovery, cybersecurity, and decentralized computing. AVALW News is our journalism platform. AVALW Shield (avalw.ai) is our on-device privacy and security product.
- Registered name
- AVALW S.R.L.
- Location
- Iasi, Romania
- Tax ID (CUI)
- 40966413
- CAEN code
- 6391 (Web portal activities)
- Founded
- 2019
- Platforms
- avalw.com, avalw.ai, avalw.org, avalw.io
Contact
We respond to all inquiries within 48 to 72 business hours.