How to Protect Your Art From AI Without Disappearing From Search Results
I’ve been thinking a lot about how artists can protect their work from AI, especially when so much of what we create is available for anyone to see online.
As an artist, I want people to find my paintings. I want them to visit my website, discover my work, and hopefully fall in love with something enough to bring it home. That’s a pretty big part of why I share my work online in the first place.
But I also don’t particularly love the idea of my artwork being collected to train an AI image generator without my permission.
So I started looking into what artists can do about it, and I learned something I think is worth sharing.
AI bots that collect content for training and AI bots that help people discover your work aren’t necessarily the same bots.
And in some cases, you can tell one to stay away while welcoming the other.
If you have your own artist website, this is something worth understanding.
Can you stop AI from using your artwork?
The short answer is that you can take steps to discourage it, but you can’t guarantee that your images will never be collected or used.
That’s probably frustrating to hear, but it’s important to understand what we’re working with.
When you put an image on a public website, people can generally save it. Someone can take a screenshot. Automated programs can collect images, too.
There are tools and settings that make some of this harder, and there are ways to communicate that you don’t want your content used for AI training. Some companies have agreed to respect those preferences.
Others may ignore them entirely.
For me, the question became how to use the controls we do have without interfering with the ability of potential customers to find our work.
Not all AI bots are doing the same thing
This was the most interesting part of my research.
When we talk about AI crawling our websites, it’s easy to lump everything together. But there are a few different things happening.
AI training bots visit websites to collect information that may be used to develop AI models. OpenAI has one called GPTBot. Anthropic, the company behind Claude, has ClaudeBot.
AI search bots help AI-powered search tools discover information on the web. These include OAI-SearchBot for ChatGPT Search and Claude-SearchBot for Claude.
User-requested bots visit web pages when someone specifically asks an AI assistant to retrieve information.
That distinction matters.
Imagine someone asks ChatGPT to recommend colorful original paintings for a Palm Beach-inspired home. I’d love for my website to be one of the places it can discover and recommend.
I would also prefer that my paintings weren’t being collected to train an image generator.
Fortunately, some AI companies let website owners express those preferences separately.
How do you tell AI training bots to stay away?
This is where a little website file called robots.txt comes in.
Most websites have one. You can usually see yours by adding /robots.txt to the end of your domain name.
For example: yourwebsite.com/robots.txt
This file contains instructions for automated programs that visit your website. Search engines have used it for years, and several AI companies now recognize specific instructions for their crawlers.
Here is a basic example of how you can ask some known AI training crawlers to stay away:
User-agent: GPTBot
Disallow: /
User-agent: ClaudeBot
Disallow: /
User-agent: Applebot-Extended
Disallow: /Those instructions communicate that you don’t want those services accessing or using your site’s content for their specified training purposes.
They don’t automatically block Google’s regular search crawler or the separate search crawlers used by ChatGPT and Claude.
There are additional crawlers you may want to consider, including Common Crawl’s CCBot, but I recommend checking the current documentation before adding rules.
And please don’t just paste these lines into an existing file without checking the rest of it. Your website may already have instructions that need to be preserved.
What if you don’t know how to edit robots.txt?
That’s perfectly fine. You don’t need to become a web developer to make an informed decision about how your work is used.
If you have someone who manages your website, ask them to review your existing crawler settings and explain whether you can restrict AI training bots while keeping your website accessible to search engines.
If you’re managing your own website, the process will depend on your platform.
WordPress websites often provide a way to manage these instructions through an SEO plugin or hosting configuration. Other website builders may offer their own settings or may limit what you can control.
Before making changes, make sure you understand what your particular platform supports.
If it helps, here is a message you can copy and send to whoever manages your website or hosting.
Hello,
I would like to keep my artwork from being used to train AI models while keeping my website visible in Google and AI-powered search. Could you check my current robots.txt file and add the instructions below, without removing anything that is already there?
User-agent: GPTBot
Disallow: /
User-agent: ClaudeBot
Disallow: /
User-agent: Applebot-Extended
Disallow: /Please make sure Googlebot, OAI-SearchBot and Claude-SearchBot can still visit my site.
I would also like to know:
1. Whether you can block specific AI bots at the server level, and which ones.
2. Whether I can see which AI bots have visited my site recently.
3. Whether anything on my account could be slowing down or blocking search engines.Thank you,
[Your name]
Does robots.txt really protect your artwork?
Here’s the catch.
A robots.txt file is essentially an honor system.
You’re publishing instructions that ask automated visitors to respect your preferences. Several major AI companies say they follow those instructions, which is certainly useful.
But nothing physically prevents a dishonest scraper from ignoring them.
This is where additional website security can help. Some hosting providers and security services can identify and block certain automated requests before they reach your content.
Even those protections have limits. A scraper can disguise itself, and someone can always save an image that’s publicly displayed in a browser.
I wouldn’t suggest spending a fortune trying to make publicly displayed artwork impossible to download. That’s a battle with some pretty significant limitations.
I would, however, take advantage of reasonable protections that don’t interfere with the purpose of having an artist website.
Should artists block AI search bots?
This is where I’d be careful.
If your goal is to sell your work, attract commissions, or build an audience, you probably want people to discover your website through as many relevant sources as possible.
And increasingly, that includes AI-powered search.
Someone might ask ChatGPT for original art recommendations, search Perplexity for a local photographer, or use Claude to research artists working in a particular style.
I want to be part of those conversations when my work is relevant.
Blocking an AI training crawler doesn’t necessarily require blocking the company’s search crawler. OpenAI and Anthropic both document separate controls for these purposes.
Google has another setting called Google-Extended. It lets website owners restrict certain uses of their content for Gemini model training and grounding. Google says this doesn’t affect normal Google Search indexing or ranking, although the broader implications for AI-powered discovery are worth considering separately.
For my own website, preserving search visibility is a priority. I’ve spent years creating content that helps people find my work and services, and I don’t want to accidentally shut off a source of traffic while trying to protect it.
That’s why I would approach crawler restrictions selectively and check the consequences of each one.
What about protecting artwork on Instagram, Etsy, and other platforms?
There’s an important difference between having your own website and sharing your work on a platform someone else owns.
On your own website, you generally have more control over crawler instructions, image sizes, hosting settings, and security.
When you upload artwork to Instagram, Etsy, Pinterest, or another service, you’re also subject to that platform’s terms and technical controls.
Changing your own website’s robots.txt file doesn’t change how those platforms handle your images.
And if the same painting is available in several places online, protecting the copy on your website doesn’t automatically protect every other copy.
This is one reason I think artists benefit from having a website they control, even if social media remains an important way of sharing their work.
A few other ways to protect your artwork online
There are a handful of practical things worth considering alongside crawler settings.
Pay attention to image sizes
You generally don’t need to upload your full-resolution original artwork files to display your paintings beautifully online. Sharing web-optimized versions can limit the resolution someone obtains from your website.
Consider watermarks selectively
They won’t prevent AI training or determined copying, but a visible signature or watermark may help communicate attribution. Whether to use one is a personal choice, especially when presentation matters.
Keep records of your original work
Save your original photographs, working files, sketches, and other documentation of your creative process. These can be useful if you ever need to establish authorship.
Ask what your website host can do
Some hosts provide bot-management tools or firewall protections. Others offer very little control. It’s worth understanding what you already have before adding more services or plugins.
And don’t put too much faith in disabling right-click. It’s an inconvenience for an ordinary visitor, not meaningful protection against automated scraping.
What I’m taking away from all of this
I don’t think artists should have to choose between protecting their creative work and being visible online.
But I also don’t think we should be promised that a simple website setting can keep our artwork out of every AI training dataset. It can’t.
What we can do is understand the difference between the automated programs visiting our websites, decide which uses we’re comfortable with, and make informed choices about the access we allow.
I plan to keep learning as these tools and policies evolve, because the technology is changing quickly and the controls available to artists are changing with it.
In the meantime, if you have your own artist website, take a look at your robots.txt file. You might be surprised to learn what your website is already telling the bots that visit it.
And if you have absolutely no idea what you’re looking at, that’s okay, too. As an artist and web designer, I spend a lot of time thinking about how creative people can make the most of their websites. This is one more piece of that conversation.
If you are deciding where else to spend your time online, I wrote about which social platforms are worth an artist’s time, and I keep a page of tools for artists. If you would like help with your own site, you can see my artist website design, and if you would like to see what I am sharing online, you can explore original work.
If this saved you a few hours of research or helped you feel a little more confident about protecting your art online, you’re welcome to buy me a coffee. I share resources like this because I think artists deserve to understand the technology affecting our work.