Fact-checking organizations are being defunded, making personal verification skills more critical than ever.

Tools to use include image verification, deepfake detection, fact-checking, media bias, and archival research.

No single tool reliably detects AI-generated content, using several together significantly improves accuracy.

The most powerful verification tool is psychological: pausing before sharing reduces misinformation's spread.

The modern news landscape is an information battlefield: shaped by AI-generated content, hyper-partisan framing, and viral rumors designed to travel faster than corrections. Digital and media literacy are no longer niche skills for journalists or academics. They are skills necessary for anyone online.

You don't need all of these tools every day. You just need to know they exist and when to reach for the right one.

Before diving in, there are two websites worth bookmarking: Bellingcat, run by a collective of investigators who both debunk viral claims and provide free open-source tools (including their AutoArchiver, which preserves online content before it's changed or removed); and Indicator, a newsletter by a professional fact-checker that covers online manipulation and hosts their own Navigator tool to help you find the right verification resource for any given claim.

Checking Images and Videos

Google Lens/Reverse Image Search—Google Lens is a great starting point for verifying images. Right-click any image in your browser or upload one at images.google.com to find where it's appeared online, check if a "breaking news" photo is actually old, or identify locations. It's a search engine, not a forensic tool, so it can't detect manipulation.

TinEye—If you want to find the oldest version of an image online, TinEye is your tool. It searches tens of billions of indexed images and excels at tracking down the original source of a photo that’s suddenly going viral. Free to use, though it only finds exact or near-exact matches, so it won’t match different photos of the same person.

InVID WeVerify—If you want to go a step further, this free browser extension (Chrome and Firefox) calls itself "a Swiss Army knife for verification". It can break videos into keyframes, run reverse image searches across multiple engines, and inspect image metadata. Some advanced features are restricted to registered journalists, but the core tools are available to anyone.

FotoForensics—FotoForensics uses a technique called "error level analysis" to highlight potential signs of digital editing. It's worth trying as one data point, but with real limitations: It works well only on certain file formats, such as JPEGs, produces frequent false positives, and is largely ineffective against AI-generated images. The developer himself says it requires expert interpretation.

Lenso.AI—(Disclosure: I was given a few free tokens to play around with this tool.) Lenso.AI is a tool for facial and object recognition that can surface visually similar images even when they've been cropped, recolored, or lightly altered. It is useful for deeper investigations that go beyond Google Lens, though the facial recognition feature is region-restricted, and full results sit behind a paywall (plans from $19.99/month).

Spotting Deepfakes and AI-Generated Content

No single tool catches everything here, but using a couple together (along with your own critical thinking) gives you a much better shot.

Hive Moderation—The free Hive Detect web tool and Chrome extension let you right-click images and text to check for AI generation. It scans for patterns from known models like Midjourney, DALL·E, and Stable Diffusion.

AI or Not—Even simpler: upload an image, audio file, or video and it tells you whether it’s likely AI-generated. Quick and easy, though free usage is limited, and the self-reported accuracy may not reflect real-world performance. Again, exercise caution, and this is just one tool during your verification journey.

GPTZero—Widely used for detecting AI-written text, but exercise real caution. It flags formal or academic writing more often than it should, misses a significant portion of actual AI content, and disproportionately flags non-native English speakers. GPTZero itself says results should inform conversations, not serve as punitive evidence.

C2PA Content Credentials—Backed by Adobe, Microsoft, Google, BBC, and hundreds of others, Content Credentials attaches cryptographically signed metadata to media. Think of it as a nutrition label showing who created something and whether AI was involved. Adobe’s free Content Authenticity web app and Chrome extension let anyone inspect Content Credentials. Not all platforms support it yet, but this is something to keep an eye on!

When a claim sounds suspicious, these resources do the heavy lifting.

Google Fact Check Explorer—A search engine for fact checks. Go to toolbox.google.com/factcheck/explorer and search any claim to see if reputable organizations have already evaluated it. It aggregates more than 150,000 fact checks. Google used to surface these in regular search results, but removed that feature in 2025, so you have to know to go looking.

Snopes—The internet’s longest-running rumor-debunking site. Great for viral hoaxes, urban legends, and those claims that sound just believable enough to share. Fact-checking sites should still always be verified, and it helps to cross-check their results.

PolitiFact—Best for political claims, using its Truth-O-Meter to rate accuracy with helpful context about why something is misleading, not just whether it’s false.

FactCheck.org—Nonpartisan and backed by the University of Pennsylvania, with a dedicated SciCheck feature for health and science claims. Especially useful right now, given how much health misinformation circulates.

The Evidence Collective—Speaking of health misinformation, I want to highlight a new science communication nonprofit that I work with, The Evidence Collective (now on Substack). Our group of scientists and clinicians writes up briefs that respond to timely health topics, in addition to sharing evidence-based health information in a variety of formats and spaces.

RumorGuard—Created by the News Literacy Project (also great!), this goes beyond labeling claims true or false. It teaches the five factors of verification (authenticity, source, evidence, context, and reasoning) using real viral claims as teaching cases. A great tool for building your own verification instincts.

AFP Fact Check/Reuters Fact Check—Especially useful for international or cross-border misinformation that U.S.-focused outlets might miss. AFP has roughly 150 fact-checkers across 26 languages.

Important note: Like AI-detection tools, fact-checking websites are imperfect. They work best when used together, over time, and alongside critical thinking, not as definitive labels for what is true or false.

Seeing Beyond Media Bias

Biased information can still have some truth. These tools help you see the full picture.

AllSides—Shows the same news story from left-, center-, and right-leaning outlets side by side. The comparison makes framing differences much more visible. Important to know: AllSides rates political lean, not factual accuracy. Like all of these media bias tools, they are a useful starting guide, not a perfect indicator.

Ground News—Highlights “blindspots,” stories covered heavily by one side of the political spectrum and barely by the other. The free version is useful for getting an overview of whether a story is covered more among right or left-wing media, but you get many more features starting at $8.33 a month.

NewsGuard—A browser extension that displays trust and reliability ratings for more than 35,000 websites as you browse. Journalists review each site against nine criteria for credibility and transparency. Free on Microsoft Edge, $4.95/month elsewhere.

Ad Fontes Media Bias Chart—A visual map ranking outlets by both political leaning and factual reliability. Widely cited and a helpful snapshot into our media ecosystem. Again, a useful starting point, not a definite conclusion.

Checking the Paper Trail

Wayback Machine—One of my favorite websites, and you can think of it as the internet’s memory. If a webpage was edited, scrubbed, or deleted, the Wayback Machine (web.archive.org) often has the original. It passed one trillion archived pages in 2025, though some major publishers now block its crawlers.

WHOIS Lookup—Want to know when a suspicious “news” site was actually created? WHOIS tools show domain age, registrar, and hosting info. Personal registrant data is now mostly redacted due to privacy regulations, but domain age alone can tell you a lot. For example, a site registered last week claiming to be an established news outlet is a red flag.

SunCalc—This is a unique tool that allows investigators to verify whether shadows in photos or videos match the sun’s position at a claimed time and location. Very cool and useful for deep-dive verification projects!

Conclusion: Slow Down and Verify

The most powerful tool in your digital detective kit isn’t any kind of software. It’s understanding our own psychology and developing the habit of slowing down before sharing. Critical ignoring—knowing what to ignore—helps us protect our attention, but when something does demand verification, this list gives you a place to start.

Disinformation doesn’t just want you to believe a lie. It wants to exhaust you, confuse you, and make truth feel not worth chasing. These tools push back against that and provide you with more agency. You don’t need to catch everything. You just need to pause, check, and decide. The more of us who do that, the less of an impact misinformation and disinformation will have on all of us.

A version of this post also appears on Misguided: The Newsletter.

There was a problem adding your email address. Please try again.

By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy

© Psychology Today