Skip to content
AEO Canon · the reference for answer-engine optimization

What HTTP Status Codes Tell an AI Crawler

HTTP status codes are instructions to AI crawlers — 200 means crawl and use, 301 consolidates signals to a new URL, 404 drops a page slowly while 410 drops it fast, 429 says back off, and repeated 5xx errors can get a page removed from the answer pool entirely.

BBurke Atkerson2 min read

To an AI crawler, an HTTP status code is a command — it decides whether your page gets used, consolidated, retried, or dropped. A 200 says use me; a 301 passes your signals forward; a 410 removes you fast. Getting these right keeps your best pages in the answer pool.

Quick answer

200 = crawl and use. 301 = follow and move your signals to the new URL. 404 = dropped slowly; 410 = dropped fast and for good. 429 and 5xx = back off and retry — but repeated server errors can remove the page from AI answers entirely.

Which status codes keep me in AI answers?

Only 200 OK actively helps — it tells the crawler the content is live and usable. A 304 Not Modified is also friendly: it confirms your cached copy is still current, saving a full fetch. Everything else either moves, delays, or removes you.

How do redirects change what a crawler keeps?

The redirect type decides which URL survives. A 301 permanent redirect tells the crawler to follow the hop and consolidate your accumulated signals onto the new URL. A 302 or 307 temporary redirect is followed too, but the crawler keeps treating the old URL as canonical. See redirects for AI crawlers for the full playbook — and avoid long chains, which burn crawl budget and can exceed a crawler's hop cap.

What removes a page from the answer pool?

Persistent errors. A single 404 or 5xx is survivable, but repeated 404s eventually drop the URL, a 410 Gone drops it faster and permanently, and repeated 500/503 responses signal an unreliable page that crawlers stop trusting.

What each HTTP status code tells an AI crawler
CodeMeaningWhat an AI crawler does
200 OKSuccessCrawls and uses the content — eligible to be cited
301 Moved PermanentlyPermanent redirectFollows and consolidates signals onto the new URL
302 / 307 Found / TemporaryTemporary redirectFollows but keeps the old URL as canonical
304 Not ModifiedUnchanged since last fetchUses its cached copy — no re-download
404 Not FoundMissing (maybe temporary)Revisits, then eventually drops the URL
410 GonePermanently removedDrops the URL faster and permanently
429 Too Many RequestsRate limitedBacks off and retries later
500 / 503 Server errorServer failureRetries later; repeated 5xx can drop the page

Serve clean 200s for what you want cited, and use 301s deliberately. More in redirects for AI crawlers and crawl budget for AI.

How do redirects affect AI crawlers?

301s pass your signals forward; 302/307 keep the old URL canonical. Avoid long chains.

Read the full answer →
What is crawl budget for AI?

The finite attention a crawler spends on your site — wasted on errors and redirect chains.

Read the full answer →
Do canonical URLs affect AI citation?

Yes — the canonical URL is the one crawlers consolidate signals onto and cite.

Read the full answer →

Frequently asked questions

What is the difference between a 404 and a 410 for AI crawlers?
Both mean the page is missing, but 410 Gone signals a permanent, intentional removal, so crawlers tend to drop the URL faster. A 404 Not Found is treated as possibly temporary, so the page is usually revisited a few times before being dropped from the answer pool.
Do redirects hurt AI crawling?
A single clean redirect is fine, but long redirect chains waste crawl budget and some crawlers cap the number of hops they will follow. If the chain exceeds that limit, the crawler gives up before reaching your real content, so keep redirects to one hop where possible.
What does a 429 status code do to an AI crawler?
A 429 Too Many Requests tells the crawler to slow down and try again later. Well-behaved bots back off and retry, but if your server keeps returning 429 or 503, the crawler may fetch you less often, which delays how quickly new content enters AI answers.

Related reading

One scannable reference for the major AI bots — GPTBot and OAI-SearchBot (OpenAI), ClaudeBot and Claude-SearchBot (Anthropic), Google-Extended, Applebot-Extended, PerplexityBot, CCBot, and Bytespider — with each bot's owner, purpose, and whether it actually obeys robots.txt.

2 min read

The Core Web Vitals thresholds are fixed — LCP good under 2.5s, INP good under 200ms, CLS good under 0.1 — and you pass only when 75% of visits hit good on all three. Speed does not directly rank you in AI answers, but it keeps crawlers fetching and users landing on the pages that get cited.

2 min read

In an XML sitemap only a few fields carry weight — loc is essential, lastmod is trusted as a freshness signal if you keep it honest, while changefreq and priority are largely ignored. Include only canonical, indexable 200 URLs and split large files with a sitemap index under 50,000 URLs.

2 min read