Should I block AI crawlers from my pricing page to stop wrong prices?

Wrong prices in AI answers usually come from somewhere other than your current pricing page. When you block crawlers from that page, you remove the correct source and leave the wrong ones in place.

As of October 2026, an engine answers in one of two ways. It either answers from what the model learned in training, or it searches the web and uses the pages it retrieves. Blocking your pricing page touches only the second path, and only for crawlers that respect the block. On that path, the engine now finds comparison sites, old reviews and forum threads, but not you. On the first path, nothing changes, because training data is already frozen.

So the fix runs the other way. Keep the page reachable, state each price in the HTML, and correct the stale copies. The playbook on fixing AI answers that get your pricing wrong covers the full process.

robots.txt rules are per bot. OpenAI's crawler documentation separates GPTBot (training), OAI-SearchBot (ChatGPT search) and ChatGPT-User (user-initiated fetches), and says each setting is independent. Sites opted out of OAI-SearchBot "will not be shown in ChatGPT search answers."

• Blocking a search bot (OAI-SearchBot, PerplexityBot, Claude-SearchBot) removes your page from that engine's index. The question still gets answered from other pages.
• Blocking a training bot (GPTBot, ClaudeBot) signals that your content should not be used for training. It does not erase what earlier models learned, and it does not affect search.
• User-triggered fetchers may not obey. OpenAI says robots.txt "may not apply" to ChatGPT-User. Perplexity says Perplexity-User "generally ignores robots.txt rules."
• Google has no robots.txt token for AI Overviews or AI Mode. Blocking Googlebot removes the page from Google Search entirely.

The training cutoff is the part blocking can never reach. OpenAI lists a February 16, 2026 knowledge cutoff for its GPT-5.6 models and August 31, 2025 for the Instant model. A model answering without search reports what it learned before then. For the wider picture, see whether blocking all AI bots hides you from AI search and the per-bot robots.txt guide.

Prices in the raw HTML

Vercel's crawler study found that ChatGPT and Claude crawlers fetch JavaScript files but don't execute them, so "they can't read client-side rendered content." A pricing table built by a JavaScript widget, or a monthly/annual toggle that loads prices from an API, can look empty to these crawlers. Googlebot renders JavaScript, so a page can look fine in Google and blank everywhere else.

A specific sentence an engine can quote

Stating a specific fact on a fetchable page is the only on-page change with a positive experimental signal. The 252,000-trial study "What Gets Cited" found that including explicit price information helped consistently. Read it carefully: the study tested pages already inside the model's context, on a synthetic corpus. It does not show that a price sentence gets you retrieved. It shows that once you are retrieved, a stated price is usable.

Say you sell project software with three tiers. A sentence like "The Team plan costs $79 per user per month, billed annually" is quotable. A price inside an image or a slider is not. Include the currency, the billing period and what the plan covers. The guide to making SaaS pricing pages readable to AI walks through the layout.

You don't want your content used for training. That is a legitimate business choice. Block the training bots and leave the search bots open:

User-agent: GPTBot
Disallow: /

User-agent: OAI-SearchBot
Allow: /

Some prices should never be quoted. Legacy rates or partner pricing can be hidden from display with nosnippet, max-snippet:0 or data-nosnippet, which Google and Bing document as on-page controls. Bing says content marked data-nosnippet stays indexed but is excluded from snippets and AI summaries. Removing stale prices from the page is usually cleaner.

You block by accident. A firewall or CDN can turn away a crawler that robots.txt allows, and nothing on your dashboard says so. Read when your CDN or firewall blocks AI crawlers before assuming access is fine.

You want AI out entirely. That is a different goal with different limits, covered in whether you can completely stop AI from using your content.

1. Check that robots.txt allows OAI-SearchBot, PerplexityBot and Claude-SearchBot on your pricing URL.
2. Request the page as each crawler and confirm it gets a 200, not a 403 or a challenge page. The AI Crawler Checker does this per bot.
3. Open the raw page source (not the browser inspector) and confirm every price appears as text.
4. Make sure the same price appears everywhere on your site. Run the check for contradicting facts on your own site.
5. Find third-party pages showing old prices and ask their owners to update them. When engines cite other sites, being named correctly there is associated with getting into the answer.

What you control is access and the facts on your pages. What the models say, and what they learned before their cutoff, you don't.

Will unblocking my pricing page fix wrong AI prices right away? No. Search-based answers can only use your page after a crawler fetches it again and the engine retrieves it for a question, and neither step is guaranteed or on a schedule you set. Answers given without search draw on training data with a fixed cutoff, so they can keep repeating old prices until a newer model ships. Unblocking removes an obstacle. It does not force a correction.

Can I block GPTBot but still appear in ChatGPT search? Yes. OpenAI states that each crawler setting is independent. You can disallow GPTBot to signal that your content should not be used for training, while allowing OAI-SearchBot so your pages can appear in ChatGPT search answers. Blocking GPTBot does not remove you from ChatGPT search, and allowing it does not get you cited there.

Does schema markup on my pricing page help AI get prices right? Not as far as the evidence shows. Google says structured data isn't required for generative AI search, and Ahrefs' matched test found AI Overviews citations fell 4.6% after pages added schema. Existing markup does no harm, so there is no need to remove it. Put the price in visible, server-rendered text, which is what crawlers that don't run JavaScript can read.

What if my prices change often? Keep one canonical pricing page and update it first, then remove or redirect older pages that list previous prices. The same 252,000-trial study reported that a recent timestamp also helped within context, but the wider evidence on freshness is unresolved. Treat the date as accuracy hygiene, not a ranking trick. Third-party copies will lag, so check them after each change.