Introducing NewsGuard AI, a reliable source of news 80%

By Jim Heaney84%

8/6/2026, 8:45:56 AM

BS Summary: This article contains 31 faulty reasoning types, including Halo Effect, Appeal to Authority, and Anecdotal, with Attempt to Sell a Product or Service as the most egregious example at 41.8% saturation with 309 hits. Analysis detected 1,962 faulty-reasoning hits from 740 analyzed words, generating a BS Score of 64% and a BS Rank of 80% (5,926 of 28,846 articles). This article is worse (more manipulative) than 79.50% of the article peer group.

You’re no doubt familiar with the phrase “garbage in, garbage out.” 
The concept can be applied to much of what you find on the Internet. 
Propaganda, disinformation and flat-out lies pass ing as news, distributed on websites and social media platforms, facilitated by artificial intelligence to a growing degree. 
It’s getting harder and harder to know what to trust. 
Enter a new service: NewsGuard AI . 
It’s produced by NewsGuard , which since 2018 has reviewed and rated some 36,000 online news and information platforms for accuracy and transparency based on nine, nonpartisan standards. 
Some 12,000 have received a passing grade. 
Content from those sites will be mined to produce AI-generated stories. 
Every fact and claim will be footnoted with a link to the publication it was drawn from. 
Subscribers can create custom searches or choose from story lines generated by NewsGuard AI. 
It functions as a personalized news service. 
I recognize that artificial intelligence has a stigma attached to it. 
It goes back to “garbage in, garbage out.” 
Most chatbots generate content from online sources regardless of their credibility. 
The result: The leading AI models spread false claims on controversial topics 35 percent of the time, according to a NewsGuard analysis. 
Stories generated by NewsGuard AI only draw from reliable sources of news and information. 
No garbage in, no garbage out. 
This field is for validation purposes and should be left unchanged. 
NewsGuard AI launched in late June and is rolling out in partnership with select outlets, including Investigative Post. 
Other inaugural launch partners include The Atlantic magazine, whose CEO praised NewsGuard AI for its “verified sources, clear citations, and a compensation model for publishers,” referring to its 50-50 revenue sharing with news publishers whose journalism is cited. 
Readers of Investigative Post  that’s you  are eligible for a free one-month trial subscription to NewsGuard AI, after which you’re eligible for a 33 percent discount off the $6 monthly subscription fee, half of which goes to Investigative Post. 
You can sign up here . 
I was turned on to NewsGuard last summer when I read The Death of Truth , an eye-opening book by Steven Brill that details, in the words of the book’s subtitle, “How Social Media and the Internet Gave Snake Oil Salesmen and Demagogues the Weapons They Needed to Destroy Trust and Polarize the World.” 
Brill, co-CEO of NewsGuard, is a journalist, entrepreneur and media watchdog. 
The other CEO is Gordon Crovitz, former publisher of The Wall Street Journal. 
They’re smart, serious people. 
Donate to support our nonprofit newsroom 
I was sufficiently impressed by the book and my subsequent interactions with Brill and his team that I invited them to make a presentation on their work at an event Investigative Post is hosting Oct. 22 at the Burchfield Penney Art Center, adjacent to Buffalo State University. 
Brill will discuss the key findings of his book, while Sofia Rubinson, a senior editor, will detail how NewsGuard goes about rating online news outlets for credibility and accuracy. 
Their presentation will also include a demonstration of NewsGuard AI , 
Tickets are only $10 and can be purchased here . 
You can learn more about NewsGuard AI by reading this press release from the company, which includes this description: 
The 12,000 news and information sites rated as reliable  are aggregated under a big tent that includes newspapers, magazines and broadcasters, opinion sites from across the political spectrum, local news startups, single-author news and commentary websites and Substacks, as well as the informational websites of state and local governments, think tanks, hospital systems, and research universities assessed as credible. 
NewsGuard journalists trained NewsGuard AI with 41 editorial safeguards designed to provide responses that cite sources for every fact, report all sides of a story even-handedly with nuance, and suggest follow-up prompts based on a reader’s interests. 
Users will also be able to search the ratings of news sites and access NewsGuard’s Reality Check daily newsletter. 
Put simply, NewsGuard AI is for people who want real news and information from reliable sources. 
Also, here’s a report from CNN , which quotes Brill as saying: “Think of how the existing AI chatbots operate, and in every respect, NewsGuard AI does the opposite.” 
I’m generally more skeptical than most people  it comes with the job  but I’m impressed with NewsGuard and its artificial intelligence chatbot. 
I encourage you to check it out. 
Article reasoning-pattern comparisonThis article: 14.6%Jim Heaney: 4.1%Investigative Post: 3.0%Confirmation Bias14.6%This article: 7.4%Jim Heaney: 1.4%Investigative Post: 1.0%Anchoring Bias7.4%This article: 4.7%Jim Heaney: 3.0%Investigative Post: 2.5%Availability Heuristic4.7%This article: 10.0%Jim Heaney: 1.1%Investigative Post: 0.6%Representativeness Heuristic10.0%This article: 0.0%Jim Heaney: 0.7%Investigative Post: 0.4%Hindsight Bias0.0%This article: 2.4%Jim Heaney: 1.9%Investigative Post: 0.6%Overconfidence Bias2.4%This article: 5.1%Jim Heaney: 6.7%Investigative Post: 3.3%Framing Effect5.1%This article: 6.9%Jim Heaney: 0.9%Investigative Post: 0.4%Loss Aversion6.9%This article: 3.8%Jim Heaney: 0.2%Investigative Post: 0.4%Status Quo Bias3.8%This article: 6.4%Jim Heaney: 0.4%Investigative Post: 0.3%Sunk Cost Effect6.4%This article: 3.2%Jim Heaney: 1.5%Investigative Post: 1.4%Optimism Bias3.2%This article: 4.6%Jim Heaney: 2.8%Investigative Post: 1.2%Pessimism Bias4.6%This article: 7.6%Jim Heaney: 12.2%Investigative Post: 9.9%Negativity Bias7.6%This article: 4.7%Jim Heaney: 1.2%Investigative Post: 0.6%Self-Serving Bias4.7%This article: 0.0%Jim Heaney: 1.2%Investigative Post: 0.8%Fundamental Attribution Error0.0%This article: 0.0%Jim Heaney: 0.0%Investigative Post: 0.1%Actor-Observer Bias0.0%This article: 0.0%Jim Heaney: 1.3%Investigative Post: 0.5%In-Group Bias0.0%This article: 0.0%Jim Heaney: 0.2%Investigative Post: 0.2%Out-Group Homogeneity Bias0.0%This article: 19.6%Jim Heaney: 2.1%Investigative Post: 0.5%Halo Effect19.6%This article: 0.0%Jim Heaney: 0.0%Investigative Post: 0.0%Horn Effect0.0%This article: 0.0%Jim Heaney: 0.0%Investigative Post: 0.0%Dunning-Kruger Effect0.0%This article: 2.4%Jim Heaney: 0.3%Investigative Post: 0.9%Recency Bias2.4%This article: 11.2%Jim Heaney: 0.7%Investigative Post: 0.2%Primacy Effect11.2%This article: 0.0%Jim Heaney: 0.0%Investigative Post: 0.0%Blind-Spot Bias0.0%This article: 0.0%Jim Heaney: 2.0%Investigative Post: 0.7%Ad Hominem0.0%This article: 0.0%Jim Heaney: 0.5%Investigative Post: 0.2%Straw Man0.0%This article: 19.1%Jim Heaney: 4.9%Investigative Post: 2.3%Appeal to Authority19.1%This article: 10.5%Jim Heaney: 1.7%Investigative Post: 0.7%False Dilemma10.5%This article: 0.0%Jim Heaney: 0.3%Investigative Post: 0.7%Slippery Slope0.0%This article: 0.0%Jim Heaney: 0.3%Investigative Post: 0.1%Circular Reasoning0.0%This article: 10.5%Jim Heaney: 5.3%Investigative Post: 2.9%Hasty Generalization10.5%This article: 0.9%Jim Heaney: 0.7%Investigative Post: 0.2%Red Herring0.9%This article: 0.0%Jim Heaney: 1.0%Investigative Post: 0.4%Bandwagon0.0%This article: 1.6%Jim Heaney: 2.8%Investigative Post: 2.6%Appeal to Emotion1.6%This article: 11.4%Jim Heaney: 1.3%Investigative Post: 0.5%Begging the Question11.4%This article: 1.4%Jim Heaney: 2.3%Investigative Post: 1.3%Post Hoc (False Cause)1.4%This article: 0.0%Jim Heaney: 0.1%Investigative Post: 0.0%Tu Quoque0.0%This article: 0.0%Jim Heaney: 0.5%Investigative Post: 0.4%Burden of Proof0.0%This article: 0.0%Jim Heaney: 0.5%Investigative Post: 0.1%Appeal to Nature0.0%This article: 8.1%Jim Heaney: 0.7%Investigative Post: 0.2%Composition/Division8.1%This article: 16.9%Jim Heaney: 3.9%Investigative Post: 1.7%Anecdotal16.9%This article: 0.8%Jim Heaney: 0.8%Investigative Post: 0.1%No True Scotsman0.8%This article: 0.0%Jim Heaney: 0.4%Investigative Post: 0.5%Ambiguity (Equivocation)0.0%This article: 0.0%Jim Heaney: 0.0%Investigative Post: 0.0%Gambler’s Fallacy0.0%This article: 0.0%Jim Heaney: 0.0%Investigative Post: 0.2%Middle Ground0.0%This article: 0.0%Jim Heaney: 0.2%Investigative Post: 0.0%Personal Incredulity0.0%This article: 0.0%Jim Heaney: 0.2%Investigative Post: 0.1%Special Pleading0.0%This article: 0.0%Jim Heaney: 0.7%Investigative Post: 0.1%Genetic Fallacy0.0%This article: 3.9%Jim Heaney: 3.4%Investigative Post: 0.9%Unattributed Quote3.9%This article: 6.9%Jim Heaney: 0.8%Investigative Post: 0.5%Quote-first Misdirection6.9%This article: 13.4%Jim Heaney: 12.2%Investigative Post: 3.0%Biased Writer Voice13.4%This article: 3.2%Jim Heaney: 1.5%Investigative Post: 0.5%Indoctrination3.2%This article: 0.0%Jim Heaney: 4.4%Investigative Post: 0.8%Politically Left Leaning Bias0.0%This article: 0.0%Jim Heaney: 0.1%Investigative Post: 0.1%Politically Right Leaning Bias0.0%This article: 41.8%Jim Heaney: 7.6%Investigative Post: 1.3%Attempt to Sell a Product or S…41.8%

740 words analyzed.

Speakers

3speakers18%attributed speech605writer words
Selected voice

CNN

100%flagged-word coverage
29 attributed words21% of attributed speech92% writer coverage
0%50.0%100.0%Unattributed Quote+100.0 ptsWriter: 0.0%CNN: 100.0%100.0%Quote-first Misdirection+100.0 ptsWriter: 0.0%CNN: 100.0%100.0%Attempt to Sell a Product -44.8 ptsWriter: 44.8%CNN: 0.0%0.0%Biased Writer Voice-11.7 ptsWriter: 11.7%CNN: 0.0%0.0%Indoctrination-4.0 ptsWriter: 4.0%CNN: 0.0%0.0%

Attribution is sentence-level. Pattern percentages are calculated only from words assigned to that voice.

Loading…
Loading…
Loading…
Loading…

Analysis

Hover over highlighted words in the article to view the associated bias or fallacy analysis.