4 AI tools to help newsrooms avoid spreading harmful content

Feb 12, 2024 in Combating Mis- and Disinformation
Circuitboard with chip reading "AI"

There is no shortage of false, harmful information online today. From conspiracies and hate speech to AI-generated disinformation, this content takes on many different forms. 

And whether spread intentionally or not, the misinformation leads to offline consequences. One report from the World Health Organization found that spreading inaccurate information during crises such as infectious outbreaks and disasters can have a negative impact on our mental health and hinder healthcare efforts. 

The emergence of AI technologies today has further complicated the ability for people to distinguish fact from fiction. It’s crucial that journalists and news consumers alike be able to do so – especially this year, with so many major elections looming globally.

Here’s a list of AI tools journalists can use to detect false information, find reliable sources and more. 


Created by former journalism student Edward Tian in January 2023, GPTZero was initially designed to detect content created using ChatGPT. It has since been expanded upon to be able to spot AI-generated content from other tools, such as Bard and LLaMa. GPTZero also allows users to scan documents for plagiarism. 

Editors can use GPTZero to better assess if the articles they receive were written by humans. Reporters can utilize the tool to determine the authenticity of information they receive from their sources. 

It’s important to note that GPTZero can occasionally return false positives, so users should always follow up and do their own diligence before discarding work outright.


  • Free version, with limited features
  • Paid options range from $10-$23/month

Factiverse AI Editor

Factiverse AI Editor is a writing aid designed to analyze whether a statement is true or not. It is a machine learning model trained on reputable news reporting and the work of fact-checkers. It performs best when applied to well-covered topics such as climate change or COVID-19.

The tool’s interface is straightforward. You simply type directly into the editor, or paste in the text you want to verify and the tool will point users to sources that can help determine whether the statement is true or false. The AI editor itself can’t independently distinguish between accurate and inaccurate information.

When analyzing text, the tool searches fact-checking databases in real time. It can be used to fact-check claims in over 40 languages. 


  • 14-day trial available 
  • Lite version (€10/month)
  • Pro version (€25/month)
  • Larger teams of at least five members can qualify for a team rate

HARO (Help A Reporter Out)

Engaging with experts is another way to help verify information and avoid spreading false information. 

HARO helps ensure the accuracy of reporting by connecting journalists with sources that have expertise or first-hand experience on a topic. These experts share their insights and quotes, which reporters can incorporate to bolster their coverage and help increase trust among their readers. 

HARO has policies in place that prevent experts from pitching their stories to journalists or acting outside of their field of expertise.  

Note: HARO recently announced it will migrate to Cision's app, Connectively. The current HARO application will be phased out in early 2024. At the time of this publication, both remain available.


  • Free 


TrollWall is an AI-powered tool designed to protect brands’ social media profiles against hateful attacks. Journalists can use the tool to learn how much harmful comments contribute to social media users’ overall interaction with their profiles.

Manual content moderation is a laborious, time-consuming process. TrollWall offers automated detection and the ability to hide toxic comments. This can free up time for organizations’ social media moderators. 

The innovative tool is powered by a combination of natural language processing and machine learning algorithms. It works in real time, which can be particularly useful during newsrooms’ livestreams.

TrollWall includes a dashboard that allows newsrooms to create blacklists of phrases, maintain keyword blocklists, and track how many comments were hidden from social media profiles. This enables journalists to gain a deeper understanding of the sources of the harmful content.


  • 14-day trial available
  • Subscriptions start at €75/month, with pricing based on the number of comments processed

Photo by Igor Omilaev on Unsplash.