Product

A claim detection tool that reads and listens across everything you monitor, surfacing the specific statements worth checking.

The Claims view surfaces what matters most: top claims ranked by SIFT score and relevance, updating in real time so your team always knows where to focus. Filter by the custom categories you define, zero in on claims not yet reviewed, or step back in time with date selection to track how narratives evolve.

Confirm the claims that need action, dismiss the ones that don't, and stay ahead of the conversation as it happens.

Relevant claims get a full SIFT analysis: an AI-generated verdict on accuracy, a harm score out of 100, and a breakdown of the factors that make a claim dangerous: how misleading it is, how believable it appears, how actionable it could be, and how consequential if believed.

Dig deeper with claim interpretations, source-linked corrections, and key facts, so your team can make fast, well-evidenced decisions on what to act on next.

Tell Full Fact AI what to watch. Add the keywords and topics that matter to your work, and it scans across RSS feeds, newspapers, YouTube, TikTok, Instagram, podcasts, social media accounts, and broadcast radio, surfacing relevant claims as they appear.

The more precisely you define your topics, the more focused your results tend to be. It all starts with what you set up here.

Connect Full Fact AI to Slack and receive a notification when a claim matches something you're tracking, with the matched claim, source, and a link to review it, all in one message.

You can star the claims that matter most and filter alerts to those only, or mute specific claims you've already dealt with, helping you focus on what's most relevant to your work.

When a claim you've already fact checked resurfaces, you'll know about it instantly.

Full Fact AI matches incoming claims against your fact checking history and fires an alert the moment a repeat is detected, so you can publish a new fact check, file a correction request, or respond on social media before the misinformation takes hold.

Alerts land in a shared review queue. Work through the potential matches as a team: confirm the genuine repeats, dismiss the false positives, and keep everyone aligned on what's been actioned.

Because most misinformation isn't new, the queue becomes more valuable over time: the more your organisation fact checks, the more matches it finds.

Tracked Claims are your fact-checked statements, plus any other claims you want to watch for repeats. They are what the Alerts and Live tools match against, so the more you add, the more useful those tools become.

The more claims you add, the more the system has to work with. Each one increases the chance of catching a repeat before it gains traction.

Search gives you a view of everything being said across the sources you monitor. Claims identified in the last 24 hours are listed in time order, grouped by article, so you can see what was published, where, and by whom.

Type any keyword to filter instantly, or open the advanced filters to narrow by date, publication, claim type, category, or speaker. The URL updates as you search, so you can share a filtered view directly with a colleague. You can also save searches to come back to later.

The advanced search gives you precise control over what you see. Filter by exact phrase, required words, or excluded terms. Extend the date range from 24 hours to a week or a month. Narrow by publication, or filter by AI-classified claim type: quantity, correlation, opinion, prediction, voting, and more.

Search by claimant name with autocomplete that merges name variants and catches references as well as direct quotes. Or filter by claimant group to see everything said by politicians of a given party across your monitored sources.

The Live tool brings together transcripts from every source type in one place: broadcast channels, radio, YouTube, TikTok, Instagram, podcasts, and on-demand content. Claims are identified as transcription runs and matches to your tracked claims appear on the right.

For live streams, the page updates automatically as new content comes in. You can also paste any text directly using Block Transcripts, useful for checking a speech or statement.

Open a source and see the full transcript alongside matched claims. Identified claims appear, and any matches against your tracked claims surface on the right, with the verdict and a link to the original fact check.

Filter by claim type to focus on what matters, or toggle Highlights Only to hide the noise and see just the flagged moments.

Accurate, automatically generated transcripts for every piece of audio and video content. Claims are highlighted inline, so you can read, search, and reference them instantly.

Add TV channels and Full Fact AI will transcribe their output as it broadcasts, scanning for claims in real time and matching them against your tracked claims the moment they are spoken.

Set up recurring transcriptions for the programmes that matter. Choose your source, the days it airs, and your timezone. Full Fact AI handles the rest, transcribing each broadcast automatically.

Each schedule records when it last ran and when the next one is due, so you can see at a glance what has been covered and what is coming up.