Start with the original file and check for verifiable provenance, such as a C2PA Content Credential when one is available. Then verify the scene and claim independently. A credential can help explain who or what created a file and how it changed; it cannot prove that the scene shown is real or accurately described. A detector’s “not detected” result is not proof of authenticity.
First separate the two questions
When assessing an image or video, ask two different things:
As an Amazon Associate I earn from qualifying purchases.
- What is this file’s recorded history? Provenance tools may show information about its origin, edits, or use of AI.
- Does it accurately depict the event or claim? That requires checking the source, context, and independent evidence.
These questions are related but not interchangeable. The Coalition for Content Provenance and Authenticity (C2PA) says Content Credentials can record provenance information in a tamper-evident structure, complementing media literacy, fact-checking, and digital forensics. They do not make a value judgment about whether the depicted scene is true. C2PA’s explainer
A practical check, from the original file to independent evidence
- Preserve the file you received. Save the original image or video if possible, rather than a screenshot, crop, or re-encoded copy. Note where and when you received it. Editing, sharing, and conversion can remove metadata or degrade detectable signals, making a derivative harder to assess. OpenAI’s Content Credentials guidance
- Check the source and caption. Find the original post or publisher and examine the account, date, location claim, and caption. Look for independent reporting or other recordings of the same event. A post may use genuine footage with a misleading date, location, or description.
- Check for Content Credentials. If a C2PA-aware verifier or application is available, inspect the credential’s signer and its recorded origin, AI-use, and edit information. Consider whether the issuer and the chain of information are credible for the question you are asking. C2PA’s verification machinery can show whether credential information is well-formed and associated with the asset without evident tampering; the assertions and signer still need interpretation. C2PA’s explainer
- Read detector results narrowly. A detected supported signal may associate a file with a supported source, but it does not establish that the depiction is accurate, unedited, owned by the poster, or presented in context. “Not detected” means the tool did not find a signal it supports—not that the file is camera-original or non-AI. OpenAI’s Content Credentials guidance
- Use visual oddities as leads, not verdicts. Inconsistent lighting, reflections, text, edges, facial details, or motion continuity can be reasons to investigate further. One oddity is not proof: compression and ordinary editing can also create artifacts, and automated systems can make mistakes.
- Seek qualified help when the stakes are high. If the media could affect someone’s safety, reputation, or a major decision, look for reliable independent corroboration and consider qualified forensic review. C2PA describes provenance as complementary to fact-checking and digital forensics. C2PA’s explainer
What Content Credentials can—and cannot—tell you
C2PA defines provenance as facts about a digital asset’s history. Content Credentials can contain assertions about origin, modifications, and AI use, cryptographically bound to an asset. Verification can help show whether those assertions are associated with the asset and whether the credential or asset has been tampered with. A valid credential is evidence about recorded file history, not a guarantee that the recorded account is complete or that the content tells the truth.
#1 Best Overall
- Used Book in Good Condition
Credentials are optional, and C2PA warns against treating media without them as inherently less trustworthy. A file may lack credentials because none were added or because they were lost during handling. Their absence alone does not show that media is fake. C2PA’s explainer
Why a detector’s result is not a final answer
Detection tools look for signals within their coverage; no result should be stretched beyond that. OpenAI says supported OpenAI-generated images use Content Credentials and SynthID, and its verification services check supported OpenAI provenance signals. Its documentation also explains that a not_detected result can occur if metadata has been stripped, a watermark degraded, or the file came through an unsupported or legacy path. The checker does not detect every other company’s AI model. OpenAI’s Content Credentials guidance
Rank #2
- Used Book in Good Condition
There is no general consumer accuracy percentage established here for determining whether any image or video is AI-generated. Results depend on the system, supported signals, file condition, and test conditions; a score without that context is not a reliable verdict. NIST’s guidance on automated analysis in identity proofing calls for testing against genuine and manipulated media, tracking false positives and false negatives, and augmenting automated decisions with manual review. That guidance concerns identity proofing, not a universal guarantee about consumer detectors. NIST SP 800-63A
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteHow to report what you found
Describe evidence precisely rather than labeling a file simply “real” or “fake.” For example: “The verifier found a credential associated with this file, signed by [issuer], recording [origin or edit information]. That supports this account of the file’s history, but it does not independently verify the event shown.” Or: “This tool did not detect a supported signal; that result does not establish whether the file is AI-generated.” Then state what independent evidence does—or does not—corroborate the claim.
Quick Recap
Best Value
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
