Exposing Dangerous Falsehoods. Restoring Truth.
Explore ADHI’s AI-powered fact-checking and narrative correction tool.
the tool
what Is authentic8?
Authentic8 is ADHI’s flagship fact-checking and narrative accountability tool. Designed to expose dangerous falsehoods and restore truth,
Authentic8 operates through an eight-step investigative cycle: Monitor → Identify → Analyse → Research → Conclude → Correct → Publish → Share.
From real-time surveillance of harmful rhetoric to delivering verified insights and counter-narratives, Authentic8 helps dismantle dehumanising discourse and ensures facts reclaim their rightful place in public dialogue.
the 8-step cycle
a breakdown of the process
The process begins with continuously scanning online platforms—news sites, social media, and fringe networks—for emerging content. AI tools like natural language processing and web scraping track trends in real time, while human analysts filter relevant data and spot early signs of risk.
AI Functions: Natural Language Processing (NLP), web scraping, data clustering.
Human Review Points: Relevance of data sources, detection of anomalies, platform-specific trends.
Real-world Example: Tracking sudden spikes in mentions of a minority group across Telegram and Twitter.
Once the system flags potential content, patterns are examined to pinpoint dangerous narratives or falsehoods. AI identifies recurring keywords and topics, but it’s the human reviewers who assess intent and distinguish harmful messages from satire or commentary.
AI Functions: Pattern recognition, keyword spotting, topic modeling.
Human Review Points: Determining intent, distinguishing satire from harmful content.
Real-world Example: Identifying a repeated phrase falsely linking immigrants to crime statistics.
Each narrative is broken down to understand its tone, emotional charge, and spread. Sentiment analysis and source credibility scores support this step, while human insight ensures the full context is captured, especially for culturally nuanced or covert messages.
AI Functions: Sentiment analysis, virality prediction, source credibility scoring.
Human Review Points: Contextual nuance, cultural sensitivity, underlying bias.
Real-world Example: Analyzing a viral Facebook post framing protestors as terrorists.
The identified narrative is then placed within a broader historical or socio-political context. AI links events and data sources, but researchers verify facts, trace origins, and assess the narrative’s deeper impact through reputable sources and cultural understanding.
AI Functions: Cross-referencing databases, timeline generation, entity linking.
Human Review Points: Fact validation, use of reputable sources, narrative framing.
Real-world Example: Tracing a conspiracy theory to its origins in a fringe subreddit.
Based on gathered evidence, the team determines whether the content qualifies as disinformation or dehumanisation. While AI may generate risk scores, humans make the final call—considering ethical implications, legal definitions, and intent.
AI Functions: Risk scoring, multi-factor content labeling.
Human Review Points: Final decision-making based on evidence, ethical considerations.
Real-world Example: Flagging a manipulated video that falsely depicts violent behavior by refugees.
To address the falsehood, counter-narratives or rebuttals are crafted. AI tools can assist with drafting, but the tone and message are refined by human editors to ensure clarity, sensitivity, and alignment with the organization’s values.
AI Functions: Language generation, narrative reframing suggestions.
Human Review Points: Tone-setting, cultural appropriateness, call-to-action clarity.
Real-world Example: Creating an explainer graphic debunking the myth that a law was passed to “ban free speech.”
Once ready, the corrected content is shared through trusted channels—social platforms, partner sites, or newsletters. AI helps tailor content for each medium, while editorial oversight ensures accuracy and alignment with the overall mission.
AI Functions: Platform optimization, A/B testing, reach prediction.
Human Review Points: Editorial review, message alignment with mission.
Real-world Example: Publishing a social media thread correcting widespread claims about a protest incident.
The final step focuses on amplifying the corrected information. Using AI-powered audience targeting and influencer mapping, the message is strategically distributed. Human teams build partnerships and mobilize networks to maximize reach and impact.
AI Functions: Audience targeting, engagement tracking, influencer recommendation.
Human Review Points: Partnership vetting, community alignment.
Real-world Example: Collaborating with journalists to promote a fact-check article on a viral hoax.
stay informed
Explore recent fact-checks
Hello world!
Welcome to WordPress. This is your first post. Edit or…
Hello world!
Welcome to Starter Templates Sites. This is your first post….
How to Sell Charity to a Skeptic
Sed arcu non odio euismod lacinia. Sit amet cursus sit…
Will Charity Ever Rule the World?
Sed arcu non odio euismod lacinia. Sit amet cursus sit…
17 Signs You Work With Donation
Sed arcu non odio euismod lacinia. Sit amet cursus sit…
The History of Donation Told
Sed arcu non odio euismod lacinia. Sit amet cursus sit…
submit a claim
send in a claim for fact-checking
important FAQs
transparency & accountability
Built on Truth.
Backed by Transparency.
Our commitment to ethical practice is reflected in every step of our verification process, from how we define misinformation to how we document our corrections and involve third-party experts for unbiased review. Here’s how we stay accountable—both to the facts and to the public we serve.
The Ethical Framework We Operate With
Our work is grounded in principles of fairness, accuracy, and harm reduction. Every narrative we flag or correct is reviewed through this lens.
We operate on a set of clearly defined ethical principles that guide every aspect of our work—from how we collect data to how we respond to misinformation. These principles emphasize accuracy, fairness, harm minimization, and cultural sensitivity.
We recognize that truth exists in complex contexts, and that interventions can have real-world consequences. That’s why our interventions are deliberate, traceable, and justified. Our ethical approach also includes policies on consent, attribution, and transparency when sourcing or altering information in public discourse.
Every Edit Logged
We maintain a transparent log of fact-checks and corrections, detailing the source, the issue identified, and the action taken.
Every correction, annotation, or flag issued by Authentic8 is logged in an internal audit trail. This digital record includes metadata such as the original content reviewed, the evidence gathered, the final decision made, and the date/time of intervention.
These logs are not hidden—they’re part of our living commitment to transparency. Where appropriate, redacted versions of these logs are made publicly accessible to ensure that our process remains open to scrutiny. This audit trail helps build institutional memory, support future learning, and uphold our users’ trust.
Verified By Real People & Subject Matter Experts
Our internal team leads the first review, followed by feedback from a rotating panel of external advisors to ensure impartiality and rigor.
Authentic8 follows a two-layered verification process. The first layer involves internal analysts—researchers, technologists, and fact-checkers trained to follow strict review protocols. The second layer draws on the expertise of an independent advisory panel composed of journalists, human rights advocates, ethicists, and academic researchers.
This hybrid approach ensures that no single voice dominates our decision-making and that every major intervention is vetted through diverse perspectives. These reviewers are rotated periodically and are not involved in day-to-day operations, preserving their objectivity.
educational tools
contribute to the cause with our educational tools
Explore our suite of interactive resources designed to sharpen your critical thinking and media literacy. From the “Spot the Dehumanisation” Quiz to the “Narrative Reframer” mini-tool and practical verification tips, these tools put the power of truth in your hands.
“spot the dehumanisation” quiz
Our volunteers are at the heart of our campaigns and initiatives.
Get Involved →
“narrative reframer” Mini tool
Join us in shaping the future of global human rights monitoring.
Join The Collective →
how to contribute to the cause
Partner with us to keep our work independent, action-oriented, and inclusive.
Donate →
get involved
an invitation to all
Your actions can bring about global change. Find your calling within our collective and start contributing meaningfully.