StripShot/Guides
How-to guides on removing AI metadata, beating platform detection, and protecting your content on Instagram, TikTok, and Pinterest.
YouTube adds its AI label from your disclosure, from C2PA metadata in the file, or from its own detection. Which labels can be changed after upload, and what YouTube says about earnings and Shorts.
Google requires the IPTC trainedAlgorithmicMedia tag on AI-generated product images, and Amazon asks for a synthetic performer keyword. A full metadata remover deletes both. What to clean and what to keep.
Etsy requires AI disclosure in the listing description and at least one unedited photo of the real item. What is allowed, what is banned, how enforcement works, and where file metadata fits.
Three checks before holiday listings and sale posts go up: clean location data from photos taken at home, keep the AI tags Google and Amazon require, and disclose AI where your platform asks.
Many AI tools write content credentials or IPTC tags into the image. How to read them, and why a missing tag is not proof a photo is real.
A live vacation post says your house is empty, and a photo's GPS data can point to where you live. Post after the trip and remove the location first.
The GPS data in the file and the house number in the frame. What Ireland's Data Protection Commission advises parents to check before posting.
A photo can carry the GPS spot where it was taken, and people you send it to may be able to see it. Apple's and Google's own settings, and how to remove the location first.
Facebook shows AI info when it finds signs an image was made with AI, often in the file itself. What triggers the label, where it appears for edited vs generated photos, and when Meta requires disclosure.
Where Sora, Runway, Veo and Pika write their fingerprints in an MP4, which platforms read them, what disappears when you strip the file, and what stays no matter what you do.
Instagram, TikTok, YouTube, Pinterest, LinkedIn and X: the current AI label on each, what triggers it (C2PA, IPTC, watermarks, detectors, disclosure), the penalties, and a dated changelog. Sourced from the platforms' own pages, updated monthly.
Windows, iPhone, Mac, ExifTool and a browser folder drop: what each one really removes from a batch of photos, and how to rename the files too.
Honest removers strip C2PA, EXIF, document properties and hidden Unicode with proof. Nobody removes SynthID pixels or statistical text marks. The breakdown, and a free tool that runs in your browser.
One brand, four watermarks. Why the pixel version survives cropping and screenshots, how the text version hides in word choice, what changed on the Gemini API, and which layer a metadata tool can remove.
The EU AI Act requires machine-readable disclosure of AI-generated content. What Article 50 means for creators, which obligations already apply, and what changed in August 2026.
Screenshots do not remove C2PA Content Credentials or EXIF data. They create a new file with your device metadata instead. The technical reason, and what to do instead.
Every AI generator embeds different metadata. DALL-E adds C2PA certificates. Stable Diffusion adds your full prompt. Firefly adds Adobe signatures. Full breakdown.
Instagram, TikTok, Pinterest, LinkedIn, X, Facebook, Discord, and Reddit - what each platform strips, what they keep server-side, and what still triggers AI labels.
EXIF data is invisible information embedded in every photo you take. It contains your exact GPS location, device serial number, timestamp, camera settings, and more.