How to use the free deepfake detector
1. Choose the right tab
Image: JPG, PNG or WEBP up to 4 MB. Video: MP4, MOV or WEBM up to 30 seconds and 50 MB. Audio: MP3, WAV, M4A, OGG or FLAC up to 25 MB.
2. Upload the best copy you have
Use the original file where you can. Screenshots, re-uploads and heavy compression remove detail that detection models rely on. For a video on social media, download or screen-record the part you want to check.
3. Read the result
For an image you get a verdict, an AI-generated score, a face-swap score and, when available, the AI generator the image most likely came from. The video and audio tabs show the status of the check for your file straight after upload.
4. Check the context
A score is one signal. Find the original source, run a reverse image search for pictures, and compare with trusted channels before you share the file or act on it.
Which deepfake checker should you use?
- A profile photo, screenshot or news-style picture: use the image tab here, or the deepfake image detection page for more on what the image checker looks for.
- A short clip of someone speaking or appearing on camera: use the video tab, or see deepfake video detection for how video checks work.
- A voice note, call recording or audio clip: use the audio tab, and read deepfake audio detection to see how DeepWall finds cloned voices in newly uploaded videos.
How to read a deepfake detection score
The image checker gives two scores from 0 to 100%: how likely the picture is to be AI-generated, and how likely it is to be a face-swap deepfake. The verdict uses the higher of the two:
- 70% or more: likely AI-generated or deepfake.
- 30% to 69%: uncertain. Look harder at the source and context.
- Under 30%: likely real, as far as the models can tell.
These are probabilities from AI models, not proof. A low score doesn't guarantee a file is authentic, and a high score is not legal evidence.
What a free deepfake detector can't tell you
- It checks one file you already have. It can't tell you whether fakes of you exist somewhere you aren't looking.
- Heavy compression, low resolution, small or turned-away faces and screenshots reduce accuracy.
- The image checker is not built to detect ordinary edits such as cropping, filters or retouching.
- Daily limits apply per visitor: 5 image checks and 2 video checks.
What happens to the files you upload
Images are analysed in memory and not saved. Video and audio uploads are held in temporary storage only while they are handled and are then deleted; abandoned uploads are removed automatically. Files sent to our detection provider for analysis are deleted by the provider after processing. We don't use your files for anything else.
From one-off checks to ongoing monitoring
A free detector answers "is this file fake?" DeepWall's deepfake detection software answers a different question: "is someone publishing fakes of me right now?" It monitors newly uploaded videos for the faces and voices you enroll and alerts you. It does not monitor standalone images or audio files.
1. Enroll your face and voice
Upload 3 to 5 photos of the person to protect, plus a 30-second voice sample. DeepWall turns them into a reference profile that new content is compared against. You can update or retrain a profile at any time.
2. Continuous scanning of new uploads
DeepWall scans newly uploaded videos on the platforms it monitors in the background, around the clock. You don't paste in links or upload files.
3. Frame-by-frame matching
Faces in each video are matched frame by frame against your enrolled photos, and the audio track is compared against your voice sample. Each match gets a confidence score.
4. Alerts and evidence
You set a confidence threshold from 0 to 100%. Matches above it trigger an alert by email, SMS, push, Slack or webhook, depending on your plan. Professional and Executive plans add a timestamped evidence PDF and takedown-request templates.
DeepWall monitors newly uploaded videos on YouTube, X, TikTok, Facebook, Instagram, Reddit and news sites with embedded video. The Starter plan covers YouTube and X; Professional and Executive include all seven sources. See pricing for the full plan comparison.
Start monitoring with a free trial
Enroll 3 to 5 photos and a 30-second voice sample, set your threshold and let DeepWall watch new videos for you. Plans start at $49 per month, with no sales call required. Every plan (Starter, Professional and Executive) includes a 30-day free trial with no credit card required. Cancel anytime.
Free deepfake detector FAQ
What is a deepfake detector?
A deepfake detector is a tool that estimates whether an image, video or audio file was generated or altered by AI. Detection models look for traces that AI generators and face-swap tools leave behind, many of which are invisible to the eye.
Is this deepfake detector free?
Yes. All three checkers are free with no sign-up. Each visitor can run 5 image checks and 2 video checks per day.
Can a deepfake detector be 100% accurate?
No. Every detector makes mistakes, and accuracy drops with compression, low resolution and small faces. Use the result as one signal alongside the source of the file and your own checks.
Can I paste a YouTube or TikTok link instead of uploading?
No. The checkers work on files you upload. Download or screen-record the part of the video you want to check, then upload it to the video tab.
Does DeepWall look for deepfakes of me automatically?
Yes, for video. DeepWall's paid monitoring scans newly uploaded videos on YouTube, X, TikTok and other sources for the faces and voices you enroll and alerts you. It does not monitor standalone images or audio files. Every plan starts with a 30-day free trial.
What should I do if a file looks fake?
Don't share it as real. Save the file, the link and a screenshot with the date, report it to the platform, and warn anyone it may be used to fool. If it shows you or your organisation, consider a takedown request.