Using AI for SEO: What It Speeds Up and What It Doesn't

AI is a fast assistant for the repetitive parts of SEO. The work still needs a person who knows the business to check it.

The tasks where it saves real time:

• Keyword clustering. Paste a few thousand keywords and ask for groups by intent. You get a usable first pass in minutes instead of an afternoon.
• Briefs and outlines. A model can turn a cluster into a draft brief with questions to answer and angles to cover.
• First drafts and rewrites. It produces a starting draft that an editor then corrects, sharpens and fills with facts only you have.
• Titles and meta descriptions. It writes variants at the right length quickly. You pick and fix.
• Audit triage. Crawl exports with thousands of rows are slow to read. AI can sort issues, group duplicates and draft a fix list.
• Small scripts and regex. It writes the filter for a log file or the formula for a spreadsheet you would otherwise search for.

The common thread: AI compresses the hours, not the judgment. A wrong price, an invented statistic or a claim your product can't back up costs more than the time saved. If you work mainly in ChatGPT, our guide to using ChatGPT and OpenAI models for SEO work covers prompts and checks in more detail.

The next move is simple. Use AI for volume tasks, and keep a human review step before anything is published or acted on.

Google's position is plain. Its guidance on generative AI content says: "Appropriate use of AI or automation is not against our guidelines." The tool that wrote a page is not the problem.

Volume without value is. Google's spam policies define scaled content abuse as "when many pages are generated for the primary purpose of manipulating search rankings and not helping users." The first example listed is "Using generative AI tools or other similar tools to generate many pages without adding value for users."

OtterlyAI tested the shortcut in 2026. It loaded two new domains with roughly 1,000 AI-generated posts each. Otterly reports that Google removed one site after about 14 days and the other after about 90, with zero manual actions in Search Console on either. The removal was silent.

The test has limits. It ran one site per arm, had no human-written control, used a low-stakes topic, and was run by a vendor that sells AI visibility tracking. It is still a clear picture of what Google's written policy describes.

What it means for you: AI drafting is fine, and one-click publishing at scale is a risk to your largest channel. Our digest of Google's guidance on AI-generated content goes through the policy line by line, and whether AI content automation tools improve AI visibility covers the volume question directly.

Getting into an AI answer happens in four stages. AI writing tools touch almost none of them.

1. Parametric. What the model knows with no web access. It is frozen at a training cutoff. Nobody changes it this quarter.
2. Access. Whether a crawler may fetch and store your page at all. You and your CDN control this.
3. Retrieval. Whether your page enters the model's context for a given query. The engine operator controls most of this.
4. Generation. Which retrieved pages the model uses and cites. The model decides.

A July 2026 critical survey of 45 GEO studies (arXiv:2607.14035) concluded that "already-retrieved content can causally alter its citation or use, but no reviewed technique shows a stable, longitudinal, cross-platform causal effect on organic discoverability or downstream behavior." In plain terms: content tricks have only been shown to matter once a page is already in front of the model.

That is why a better draft doesn't fix a missing citation if the crawler never saw the page. The levers with evidence behind them sit at the access stage, and they work by removing blocks. Our guide to SEO for AI search explains each stage in more depth, and what AI SEO means separates the two senses of the term this page straddles.

Here is the split, as of October 2026.

Two items catch teams off guard. First, Cloudflare changed its defaults on 2026-09-15: Training and Agent crawlers are blocked by default on ad-displaying pages for new domains, new sites on existing accounts and existing free-tier customers. A site can lose AI crawler access without anyone editing it. Second, Google's Search Console has a generative AI inclusion control that is on by default and can be inherited from a parent property.

The practical move: treat access as a checklist you can finish, and treat retrieval and generation as things you can observe but not set.

Use AI for speed and spend your own attention on the gates. In this order:

1. Check robots.txt per bot. Make sure retrieval bots such as OAI-SearchBot, PerplexityBot and Claude-SearchBot are allowed. Training bots are a separate decision. A valid example that allows ChatGPT search while opting out of training:

User-agent: OAI-SearchBot
Allow: /

User-agent: GPTBot
Disallow: /

2. Test the firewall. Request a page as each AI crawler and confirm it gets a 200, not a 403 or a challenge page. The AI Crawler Checker does this and shows which crawlers your firewall turns away.
3. Diff raw HTML against the rendered page. Fetch with JavaScript off and compare it with what a browser shows. Prices, product names and key claims that only appear after JavaScript runs may be invisible to most AI crawlers.
4. Look for accidental suppression. Search templates for noindex, nosnippet, max-snippet:0 and data-nosnippet.
5. Confirm index status engine by engine. Check Google Search Console and Bing Webmaster Tools separately. One being healthy tells you nothing about the other.
6. State specific facts in fetchable HTML. A clear price, number or one-line definition gives an engine a sentence it can quote. Use AI to draft these, then verify every figure.
7. Then use AI for content volume tasks, with an editor on every page.

Before and after, the free checker on this site shows whether AI engines name you for up to three buyer questions. It is a snapshot, not continuous tracking.

Most mistakes come from treating AI output, or old GEO advice, as fact.

• Mass-publishing AI pages. It matches Google's definition of scaled content abuse, and in Otterly's test the removal came with no notice.
• Adding schema to get cited. Google says structured data isn't required for generative AI search. Ahrefs' matched-control test of 1,885 pages found AI Overview citations fell 4.6% after schema was added. Existing markup does no harm, but it is not a lever for AI answers.
• Publishing an llms.txt and expecting results. Google says Search ignores it, and no other engine documents reading it.
• Rewriting pages in an "AI-friendly" style. Google says you don't need to write in a specific way for generative AI search, and it rejects chunking as a requirement.
• Allowing GPTBot so ChatGPT can find you. GPTBot governs training only. ChatGPT search uses OAI-SearchBot.
• Trusting an AI audit without checking. A model summarising your crawl export can invent issues or miss real ones. Spot-check the raw data.
• Letting AI state facts about your brand. Models get specific facts wrong. Every price, date and feature in an AI draft needs a human check.

Each of these wastes effort on something the evidence doesn't support, while the access gates stay unchecked.

AI visibility is not one number. Measure each engine on its own.

In Otterly's experiment, Microsoft Copilot produced 2,718 of the 3,934 citations tracked for one site after Google had dropped it, while ChatGPT, Gemini, AI Overviews and AI Mode almost never cited it. Copilot grounds on Bing, and Bing kept the site indexed. A blended score would have hidden that split.

The first-party reports are limited. Google's generative AI report in Search Console shows impressions for AI Overviews and AI Mode, with no clicks, no click-through rate and no position, as of October 2026. Our playbook on Search Console's generative AI report covers how to read it. Bing Webmaster Tools' AI Performance report shows grounding queries, the phrases the AI actually used when retrieving content, which no prompt tracker can see.

Be wary of any tool that reports an AI rank or position in Google. Google exposes no position data to check such a number against. Track presence, citations and factual accuracy per engine, and note whether each answer used live search.

Does Google penalize content written with AI? Not for being AI-written. Google's guidance says appropriate use of AI or automation is not against its guidelines. It acts against scaled content abuse, meaning many pages generated mainly to manipulate rankings without helping users. A useful page drafted with AI and checked by an editor is fine. A thousand thin pages published overnight is the pattern Google's spam policy describes.

Will AI-generated content get my site cited in ChatGPT? Content alone does not decide it. ChatGPT search can only use pages its crawler, OAI-SearchBot, is allowed to fetch and can actually reach through your firewall. The page's key facts also need to be in the raw HTML. Once a page is reachable, retrieval and citation are decided by the engine. See whether you [need to rank on Google to show up in ChatGPT](/questions/need-google-ranking-for-chatgpt) for how the indexes differ.

Can AI tools run a technical SEO audit for me? They can speed one up. AI is good at sorting crawl exports, grouping duplicate issues and drafting a fix list. It cannot see your firewall, your Search Console settings or whether a crawler got a 403. Run those checks directly, then use AI to summarise the results. Always spot-check its conclusions against the raw data before acting.

Is it worth blocking GPTBot? That is a training decision, not a visibility one. OpenAI says GPTBot is used to crawl content that may be used to train its models, and that each crawler setting is independent. Blocking GPTBot does not remove you from ChatGPT search, which uses OAI-SearchBot. Nothing published measures whether training inclusion changes what models say about a brand.

Does AI change which keywords I should target? It changes how fast you can research them, not what engines reward. Google says you don't need to worry about covering every long-tail variation for its AI features. AI engines also rewrite a user's question into their own search queries, which you cannot observe outside Bing Webmaster Tools. Target topics your buyers care about and cover them well.