FlareFact FAQ

FlareFact is a platform and a community, not just a single tool. Right now, our main feature uses multiple AI models to read posts, links, or pasted text, extract factual claims, and check them against trusted sources. This is our core functionality for now, but we're already developing additional capabilities that will broaden the platform even further. At the moment, the app provides a clear summary, a fact-check, and an explanation for each claim—with sources included.

FlareFact uses multiple AI models to analyze content, identify factual claims, and verify them with trusted sources. It then delivers a sourced summary, fact-check, and explanation for every claim.

We understand the impact AI can have on both the planet and humanity. And while big corporations and governments continue to push it forward, we believe there's something meaningful we can do right now: give AI a positive purpose.

That's why we created FlareFact.

AI fact-checking is just the beginning—a first step toward a bigger mission. We're building a community of curious, dedicated fact-checkers committed to exploring, questioning, and pushing back against the wave of misinformation that AI can amplify when misused. Together, we're turning curiosity into action.

We don't just believe—we know that real power lies in the communities we're part of every day. That's why we're building a community of fact-checkers and developing a more comprehensive fact-checking platform to support them, free from powerful interests.

Our community brings together people dedicated to fighting misinformation and helping others make informed decisions. And unlike community-note systems tied to platforms like X or Meta, our approach is fully platform-agnostic.

You can join us at Discord.

Paid tools in FlareFact are here to fund the global initiative of community notes that will be integratable to all platforms that need them. By subscribing to our paid plans, you're directly supporting the development and maintenance of a platform-agnostic community notes system that can be integrated across various platforms, helping fight misinformation on a global scale.

Copy any kind of text, select a post from your feed or a link to an article and paste it into FlareFact. App will analyze the content and display the results, including a fact check and explanation.

Yes! Each fact check includes clickable source icons or links, so you can review the original references yourself.

AI models aren't perfect, but they potentially offer better accuracy and reliability than human fact-checkers alone when we look at the time it takes for them to analyze the content, not to mention the bias that us, humans, inherently have. To address potential bias in AI models themselves, we use multiple models developed by different groups and companies, each bringing unique perspectives and training. This diversity provides a more balanced analysis than relying on a single model, and all sources are included so you can verify the information yourself.

Post: "January 6 officer calls Trump 'petty' for Republican refusal to hang Capitol plaque…"
Fact Check: The officer's statement is subjective (opinion). It is true that Republicans have stalled the installation of the plaque. The claim about Trump letting out people who committed Jan 6 crimes is not confirmed.
Explanation: FlareFact breaks down the post into individual claims, labels each as subjective or true, and provides a summary with source links.
Sources: App will additionally display sources used in the process.

Post: "The slamming of Joe Biden is just another distraction, stop feeding into it, that's what they want."
Fact Check: The statement is subjective and reflects an opinion.
Explanation: FlareFact identifies the post as an opinion and clarifies that it cannot be verified as true or false.
Sources: App will additionally display sources used in the process.

The content may be too short, too long, lack factual claims, or be in an unsupported language. Try providing more context or splitting the content into multiple verifications if it exceeds the maximum length for a single verification. If nothing works, please contact our team.