Can you stop people downloading images from your website?
Not completely. To show a picture, a browser has to receive the picture, and whoever is in front of the screen can always take a screenshot. Any service that promises downloads are “impossible” is overselling.
What you can do is make the easy ways fail, and make the hard ways leave a trail:
- Stop the easy copying: Right-click and save, drag-and-drop, opening the image address, hotlinking from other sites, and bulk downloads by bots and AI scrapers.
- Limit what a copy is worth: Publish display-sized files, never your full-resolution original.
- Make copies traceable and prove ownership: A copyright notice, metadata, an invisible watermark and a dated record of your file.
- Find copies and get them removed: Reverse image search and takedown notices.
Every method, compared
| Method | What it stops | What it doesn’t stop | Effort |
|---|---|---|---|
| Disable right-click | Casual right-click and save | Drag-and-drop, the image address, developer tools, screenshots, bots | Minutes, free |
| Visible watermark | Clean reuse; your name stays on copies | Cropping, AI watermark removers; it also spoils the view | Per image, free |
| Smaller, lower-quality files | Printing or selling your full-size work | Using the copy online | Per image, free |
| Hotlink protection | Other sites showing your files from your server | Anyone saving the file | Server or CDN setting |
| Keeping images out of image search | Being found and copied through Google Images | Visitors to your site; crawlers that ignore the rules | Minutes, free |
| Copyright notice and metadata | “I didn’t know whose it was” | Copying; most platforms strip metadata | Minutes, free |
| Copyright registration (US) | Nothing, but it unlocks stronger legal remedies | Copying | A fee per filing |
| Reverse image search and takedowns | Copies staying up once you find them | Copies on private pages; new copies | Ongoing |
| Domain-locked, scrambled delivery | Saving, hotlinking, direct links, image search and scraper bots get a scrambled copy | Screenshots; a programmer imitating a visitor | Minutes, paid |
The rest of this guide takes them one at a time.
Disabling right-click
Most website builders have a setting or plugin for this, and on your own site it’s a few lines:
<script>
document.addEventListener('contextmenu', (e) => {
if (e.target.tagName === 'IMG') e.preventDefault();
});
</script>
<style>
img { -webkit-user-drag: none; user-select: none; }
</style>What it stops
The casual visitor who right-clicks and picks “Save image as”.
What it doesn’t
Anyone who drags the image to their desktop in some browsers, opens the image address, uses the browser’s developer tools or “Save page as”, takes a screenshot, or runs a bot. It also annoys visitors who right-click for other reasons, such as opening a link in a new tab.
Worth it? As a speed bump, yes. As protection, no: the file has already been sent to the visitor’s computer, and the image address still works anywhere.
Visible watermarks
A logo or your name across the picture, ideally over the subject, not in a corner where it can be cropped off.
What it stops
Clean reuse. Anyone who reposts it reposts your name with it.
What it doesn’t
Cropping, cloning it out, and AI tools that remove watermarks in seconds. A heavy watermark also spoils the picture for the people you want to see it, including buyers.
Worth it? For proofs and previews sent to clients, yes. For a portfolio, it’s a trade-off between protection and how your work looks.
Smaller, lower-quality files
Publish images only as large as your pages show them. Around 1,200 to 2,000 pixels on the long edge looks sharp on most screens, and is too small for a good large print.
The “72 DPI” setting you’ll often see recommended does nothing on the web. Screens ignore it; only the pixel dimensions matter. Resize the image itself.
What it stops
Someone printing or selling your full-resolution work.
What it doesn’t
The copy being used online, which is where most copies end up.
Worth it? Always. Never upload your original to a public website. Keep it on your own computer and in a backup.
Stopping other sites from showing your images (hotlinking)
Hotlinking is when another site shows your image by pointing at the file on your server, so you
pay for the bandwidth. Hotlink protection is a server or CDN rule that refuses image requests
coming from other websites. Cloudflare has a one-click setting, and on Apache or nginx it’s a
short rule that checks the Referer header.
What it stops
Other sites embedding your files.
What it doesn’t
Anyone saving the image and uploading their own copy. Requests that send no
Referer usually have to be allowed, or some of your own visitors would see broken images.
Keeping images out of Google Images and AI training
Image search is one of the main ways pictures get found and copied. To keep them out:
- Block image crawlers in robots.txt. Add
User-agent: Googlebot-ImagewithDisallow: /to keep your images out of Google Images while your pages still rank. - Block AI crawlers you know of. Add rules for crawlers such as
GPTBot,CCBotandGoogle-Extended, which train AI models. The list changes, so check it now and then. - Send “no AI” signals with your images. The
noaiandnoimageairobots tags and a text-and-data-mining reservation (TDMRep) state that you don’t allow training on your work.
What it doesn’t stop
All of this is a request. Reputable crawlers honour it; others ignore it. And it does nothing about a person saving the picture from your page.
The trade-off
Images kept out of image search can’t bring you visitors from image search either. Keep out the ones that matter and leave the rest findable.
Proving the images are yours
None of these prevent copying, but they make it much easier to get a copy taken down:
- Put a copyright notice on your site and in your image captions, such as “© 2026 Your Name. All rights reserved.”
- Add your name and copyright to each file’s metadata (EXIF and IPTC), which Lightroom, Photoshop and most photo apps can do on export. Instagram, Facebook and most marketplaces remove it, so don’t rely on it alone.
- Keep your originals and drafts. The camera raw file or layered working file is strong evidence that you made the picture.
- In the US, consider registering your copyright. Your work is protected without it, but registering before an infringement, or within three months of publishing, makes statutory damages and legal fees available. Photographers can register many published photos in one filing.
This isn’t legal advice. For a specific case, ask a lawyer.
Finding copies and getting them taken down
Search for your own pictures every so often with Google Lens and TinEye. Both find re-uploads, including resized copies.
When you find one, most sites have a copyright form or an address for takedown notices. In the US this is a DMCA notice: it identifies your work and the copy, and states under penalty of perjury that you own it and didn’t allow the use. Facebook, Instagram, Pinterest, Etsy, Amazon and the big hosting companies act on them quickly.
What it doesn’t reach
Private groups, most social posts and members-only sites never show up in search, and a copy taken down can be uploaded again.
What it doesn’t stop
Screenshots, which no website can stop. Large views carry an invisible watermark unique to each viewer, so a screenshot you find can be traced to the visit it came from, though a determined person can remove it. A programmer who writes a script imitating a visitor on your site can also rebuild the image. The full list of what it can’t do is public.
It works on WordPress, Shopify, Squarespace, Webflow, Wix and hand-built sites. There’s a 14-day free trial with no card needed, and the features page covers patrols for copies, takedown notices and dated upload records.
If you post on Instagram, Etsy or other platforms
Social networks and marketplaces make their own copy of anything you upload, and serve it from their servers. Nothing on your website can protect that copy. What you can do:
- Post a smaller copy, never your original, sized for how the platform shows it.
- Put a visible watermark or your handle on it if reposting without credit is your main worry.
- Link to your own site for the full view, where you control how the image is served.
- Search for your images now and then and use the platform’s copyright form for copies you find.
A sensible setup
For a portfolio or gallery
Publish display-sized files, add your copyright notice, keep the images out of image search if being found there doesn’t matter to you, and use domain-locked images for the work you care about most.
For an online shop
Product photos you want found in Google Images should stay findable. Protect the photos competitors copy, and use hotlink protection to stop them using your bandwidth.
Frequently asked questions
Can you stop someone taking a screenshot of your images? No. Nothing on a website can stop a screenshot. You can make screenshots less valuable (smaller files, watermarks) and traceable (an invisible watermark per viewer).
Does disabling right-click protect images? Only from the most casual visitors. The image has already been sent to their browser, and it can be saved in several other ways.
Is a watermark enough? A visible watermark keeps your name on casual reposts, but it can be cropped or removed with AI tools, and it changes how your work looks. Combine it with smaller files and monitoring.
Can I stop AI companies training on my images? Partly. Robots.txt rules and “no AI” signals are honoured by crawlers that follow the rules. Serving only scrambled copies to anything that isn’t a real visitor on your site doesn’t rely on the crawler’s good behaviour, but nothing reaches copies collected before you started.
Do I need to register my copyright? Your work is protected as soon as you create it. In the US, registering lets you sue and claim statutory damages, which is why professionals register their published work. Other countries work differently.