03/28/2024

NewsGuard and SafetyKit Launch Automated Misinformation Detection Tool

Content Moderation and Trust and Safety Teams Can Now Flag False Content at Scale Using NewsGuard’s Misinformation Fingerprints Library Paired With SafetyKit’s Powerful AI

(March 28, 2024 — New York) NewsGuard, the leading provider of human-curated trust data, and SafetyKit, the only AI automation platform built for trust and safety teams, have launched Scaled Misinformation Reviews, a product that combines NewsGuard’s human intelligence with SafetyKit Trust and Safety AI to flag misinformation at scale.

Platforms can use Scaled Misinformation Reviews to automatically find, triage, and address misinformation on their platform without spinning up a team of misinformation experts and machine learning engineers.

How it works:

  1. NewsGuard’s team of journalists curate a comprehensive, constantly updated library of the top false narratives circulating online — called the “Misinformation Fingerprints.”
  2. SafetyKit offers a customizable AI-powered engine that automatically flags content that runs afoul of a platform’s Trust and Safety policies and provides detailed policy-backed explanations for each decision.
  3. SafetyKit’s specialized Trust and SafetyAI matches content against NewsGuard’s Misinformation Fingerprints, safely distinguishing between content that advances a false claim and content that debunks that claim.
  4. SafetyKit provides platforms with a detailed Misinformation Fingerprint report containing:
    a) A detailed explanation for why this content matches the known misinformation
    b) A link to the NewsGuard Misinformation Fingerprint
    c) Customizable platform-specific suggested actions like hiding, downranking, or adding a content warning

The joint offering enables platforms to act efficiently and decisively to flag false content at scale using SafetyKit’s Trust and Safety AI — with the confidence that decisions are grounded in NewsGuard’s expert human judgment.

“We’ve seen companies try and fail to build algorithms that can flag misinformation without humans,” said Steve Brill, NewsGuard co-CEO. “With SafetyKit, we’ve finally cracked the code for how algorithms can automate detection of misinformation responsibly: By using our human-curated catalog of false claims as the data seeds that AI can reference as it classifies content.”

“Using AI to scale human intelligence is the future of Trust of Safety,” said David Graunke, SafetyKit CEO. “Combining Newsguard’s trusted human intelligence with SafetyKit’s Trust and Safety AI means platforms can review 100% of their content with human-level nuance. That means teams can fight misinformation across their platforms while avoiding costly false positives.”

Example:

NewsGuard’s team manually debunks false narratives and adds them to its Misinformation Fingerprints database, such as in the example Fingerprint below:

SafetyKit automatically flags content that matches any Fingerprints in the NewsGuard library and SafetyKit provides an explanation for why the piece of content matched NewsGuard’s Fingerprint:

This solution works across formats, topics, and languages:

  • Formats: The solution can flag misinformation in text, video, audio, and image content.
  • Topics: NewsGuard tracks false narratives across a range of topics, including health, COVID-19, elections, state-sponsored propaganda, war misinformation, climate change, and more.
  • Languages: SafetyKit can review content and flag misinformation in 133 languages.

To learn more about the solution, contact sarah.brandt@newsguardtech.com or alex@safetykit.com.

 

About NewsGuard

Founded by media entrepreneur and award-winning journalist Steven Brill and former Wall Street Journal publisher Gordon Crovitz, NewsGuard provides transparent tools to counter misinformation for readers, brands, and democracies. Since launching in 2018, its global staff of trained journalists and information specialists has collected, updated, and deployed more than 6.9 million data points on more than 35,000 news and information sources, and cataloged and tracked all of the top false narratives spreading online.

NewsGuard’s analysts, powered by multiple AI tools, operate the trust industry’s largest and most accountable dataset on news. These data are deployed to fine-tune and provide guardrails for generative AI models, enable brands to advertise on quality news sites and avoid propaganda or hoax sites, provide media literacy guidance for individuals, and support democratic governments in countering hostile disinformation operations targeting their citizens.

Among other indicators of the scale of its operations is that NewsGuard’s apolitical and transparent criteria have been applied by its analysts to rate news sources accounting for 95% of online engagement with news across nine countries.

  

About SafetyKit

SafetyKit applies Trust and Safety policies at scale with human-level nuance. SafetyKit’s founders built Trust and Safety at Stripe and Airbnb by scaling human intelligence. SafetyKit uses AI to scale even further. SafetyKit’s prebuilt integrations and no-code tools mean Trust and Safety teams can keep their platforms safe without needing engineering and machine learning help. Companies like Substack and Lime rely on SafetyKit to keep their users and platforms safe.