About Clad
Clad is an independent publication that fact-checks U.S. political news. We watch broadcast segments, panel shows, and reporting from across the spectrum, then publish a graded report card: an overall letter grade for accuracy, a factuality score, a measured read of political lean, and a plain-language explanation of why a segment earned its grade. Every report links back to its original source so you can verify it yourself. Clad is published by CladFacts LLC.
Because the value of any rating depends entirely on the method behind it, here is exactly how each part of a report is produced — and where the limits are. For the step-by-step methodology, see how Clad works. New to reading the news this way? Start with New to the news? — six short explainers on grades, lean scores, and spotting spin — or the student’s guide.
Method How we assess political leaning
Each report assigns a political-lean score on a scale from -100 (strongly left) through 0 (centered, no discernible slant) to +100 (strongly right). The score reflects only the segment under review and is based on observable cues such as word choice, which facts are emphasized or omitted, guest and source selection, and overall framing. The same criteria apply to every network and figure. Non-political content receives a score of 0. The final value is shown with both its numeric magnitude and a short rationale that names the specific cues observed in the segment.
Method How we grade for accuracy
Accuracy is reported with both a letter grade (A+ through F) and a factuality score (0–100). The letter grade evaluates severity rather than simply counting issues: A indicates high accuracy, sound sourcing, and balance with minimal problems; B indicates mostly accurate material with only minor concerns; C indicates a mix of accurate and problematic claims along with noticeable patterns such as one-sided sourcing or missing context; D indicates significant factual issues, heavy partisan framing, or major unsourced claims; F indicates pervasive misinformation or propaganda-level distortion. The 0–100 factuality score complements the letter grade by placing the segment on a continuum from entirely false (0) to entirely accurate (100).
Method How we write the summary, analysis, and key moments
The summary is a factual, non-editorial description of the segment’s content. It notes the topics covered, the sourcing approach (named versus anonymous), notable guests, and the main throughlines. The assessment section explains what context may be missing, how loaded language or selective statistics could shape viewer perception, how individual claims hold up, and any recurring patterns. Adjectives in the assessment describe the evidence itself (documented, unsubstantiated) rather than individuals. Key moments isolate three to five load-bearing claims and label each one verified, disputed, missing context, or unsupported, accompanied by a one-sentence rationale citing the relevant source or reasoning.
Method How we verify and source every report
Reports cover U.S. political news and commentary video segments from established network newscasts, morning shows, roundtables, and opinion programs, plus occasional notable clips from independent creators and specialist channels submitted by the editor — all graded by the same criteria. Transcripts are taken from on-screen captions when available; otherwise the specific video is researched via web search. Before any grade is assigned, every factual claim is checked against current authoritative sources through live web search. The system does not label a claim false simply because it is unfamiliar or recent; verification occurs first, and search results take precedence over prior assumptions. All sources cited are primary documents, official data, statutes, named experts, or reputable outlets across the spectrum. Every report also links directly to its original source video.
Method Human oversight, corrections, and your right to challenge us
Reports are drafted with AI assistance and then reviewed and approved by a human editor before publication. Corrections appear as new posts rather than silent edits to the original report, and every one is logged on the public corrections page. Readers may flag any grade or political-lean score they believe is incorrect. When that happens, the report is re-run through the exact same grading process that produced it. If the fresh result confirms the original was off, the report is removed. If it holds up, it stays — a reader disagreeing with a grade that is accurate and well sourced is not, by itself, a reason to change it.
Limits What we do not claim
This process is AI-assisted analysis of video segments. Transcripts may contain paraphrase or transcription errors. Grades evaluate the reporting and the evidence presented, not the character of any person named. The output constitutes editorial commentary and analysis; it is not legal or financial advice.
The editor
Every report on Clad is reviewed and approved before publication by Ben Cody, CladFacts' founding editor. Ben built Clad on a simple frustration: young readers are handed an endless feed with no way to tell solid reporting from spin. He sets the grading rubric, approves or rejects every AI-drafted report card, rules on reader disputes, and signs every correction. Reach him at support@cladfacts.com.
Who runs Clad & how it’s funded
CladFacts is a single-editor publication operated by CladFacts LLC, a Georgia limited liability company. No advertising, no sponsors, no investors, and no funding from any platform, party, or interest group. Every grade is free with a free account, and no grade is ever for sale. Reporting is AI-assisted and human-reviewed; the full methodology is documented on the how Clad works page.
Contact
Questions, tips, or corrections? Email the editor at support@cladfacts.com.
Clad's reports are editorial commentary, opinion, and analysis based on publicly available material, presented under fair use for commentary and criticism. Grades and verdicts assess the evidence and reporting, not the character of any person named.