ChatGPT gets its information from four places:
- Data it was trained on before release
- Live web searches it runs when it needs current facts
- Content from publishers OpenAI pays to license
- Your own context, i.e., your prompt, files, memory, and connected apps
Which one it uses depends on your question.
|
Can a website influence it? |
|||
|---|---|---|---|
|
Public web content, partner data, human trainers, synthetic data |
Always: it’s the model’s base knowledge |
Indirectly, over the long term |
|
|
Searches ChatGPT writes and sends to search providers, plus OpenAI’s own crawler |
When a question “would benefit from current information,” or when you choose Search |
Yes, by ranking for the topic in the search results ChatGPT pulls from, and by letting OpenAI’s search crawler (OAI-SearchBot) access your site |
|
|
Licensed publisher content |
Summaries of articles from publishers with OpenAI deals |
Alongside search results, for news, reviews, recipes, and more |
Only through coverage in those publications |
|
Your prompt, uploaded files, memory, chat history, connected apps |
Whenever it’s available |
Only if the user brings you in: a link they paste, a file they upload, or your app if they connect it |
Source 1: Training data
Before a model like the ones behind ChatGPT is released, it’s trained on an enormous amount of information.
That training is where ChatGPT’s general knowledge comes from: how language works, facts about the world, how to write code, what things look like, and so on.
The training happens in stages: preparing data, pre-training, post-training, and ongoing evaluation. (We go deeper on this in how answer engines work.) First comes pre-training on that huge mix of data. Then comes post-training, where human trainers and researchers (and increasingly the models themselves) teach it to follow instructions, reason, and behave safely.
What data does OpenAI use?
- Information that is publicly available on the internet
- Information that OpenAI accesses through partnerships with third parties
- Information that users, human trainers, and researchers provide or generate
What specifically is in this mix? The last time OpenAI published a detailed breakdown of its training data was for GPT-3, back in 2020. That mix was:
- Filtered Common Crawl (a huge public scrape of the web): about 60% of the training mix
- WebText2 (web pages linked from Reddit posts with enough upvotes): about 22%
- Two book collections (Books1 and Books2): about 16% combined
- Wikipedia: about 3%
Since then, OpenAI has shared less and less. Its latest system card says only that the models “were trained on diverse datasets.” So nobody outside OpenAI knows exactly what’s in ChatGPT’s training data today.
What does that mean for you? If your site is public and doesn’t block OpenAI’s training crawler, there’s a good chance your pages end up in the data future models learn from. That’s how ChatGPT can “know” your brand without searching for it. But it’s slow and indirect: your content only shows up in the next model, months later, and you can’t see what it learned.
OpenAI doesn’t use every piece of information it can gather. For example, OpenAI says that it doesn’t intentionally gather data “from sources known to be behind paywalls or from the dark web,” and that it filters out material like hate speech, adult content, sites that aggregate personal information, and spam.
OpenAI also says that they’re “increasingly” using synthetic data, training material its own models generate. OpenAI’s examples include practice prompts and examples in languages with little training text. The original ChatGPT search model was partly trained on answers written by a stronger OpenAI model (o1-preview).
Can you stop ChatGPT from training on your content?
Yes. Some site owners don’t want their work used to train AI for free, and big publishers often block it to keep leverage in licensing deals. In fact, GPTBot is the most-blocked AI crawler, according to our study of ~140 million websites.
The trade-off is that future models learn less about you from your own site. If you want ChatGPT to know your brand, you probably want to leave it open.
If you do want to opt out, there are two ways:
- If you run a website: block GPTBot in your robots.txt file. GPTBot is the crawler OpenAI uses to collect training data. Blocking it doesn’t remove you from ChatGPT’s search results, which are controlled by a different crawler, OAI-SearchBot (more on that below). The two settings are independent, though if you allow both, OpenAI says it may use a single crawl for both purposes.
- If you use ChatGPT: turn off “Improve the model for everyone” in your data controls. OpenAI says content from Business, Enterprise, Edu, and ChatGPT for Healthcare workspaces isn’t used for training by default.