Ahrefs survey: 93% of 301 marketers use AI for content

Ahrefs published its second annual survey on how marketers use AI for content on 30 September 2026, written by Glen Allsopp. It reports 301 responses. This report restates the findings, each with the base Ahrefs gives, so that no figure is read against the wrong group.

Ahrefs is a vendor of SEO and AI marketing software, and its post carries promotion for its own AI platform. The survey is a vendor’s data. The limits are set out below, and the whole of it is Ahrefs’ reporting, not ours.

For Google’s own position on AI-written content, see our report on its 1 October 2026 guidance update.

What did Ahrefs find?#

The table lists the headline figures. Each is Ahrefs’ number, and the last column is the group it applies to.

Finding	Result	Group
Use AI to create or help create content	93% (280 of 301)	All respondents
View of AI content became more positive	61%	All respondents
Produce content faster with AI	85%	AI users
Trust AI content only if a human reviews it	66%	All respondents
Publish a larger share of AI content than a year ago	69%	All respondents
Say human writing beats their AI content	56%	AI users
Use AI to check AI-assisted content	28.6% (14% in 2025)	AI users
Disclose AI use to visitors	22% (16% in 2025)	AI users
Publish more than 75% pure AI content	7% (4% in 2025)	AI users

How many marketers use AI?#

Of 301 responses, 280 (93%) said they use AI to create or help create content, up from 87% in last year’s survey. Ahrefs adds that last year had more respondents, so it compares only at a high level.

Among the 6.98% who do not use AI, Ahrefs says almost all cited accuracy as a concern and more than half said they generally distrust AI-generated content.

On sentiment, 61% say their view of AI-generated content has become more positive over the past year: 38% say somewhat more positive and 23% say much more positive. Ahrefs says it has no data on why, and suggests better models, operational efficiencies and better systems as likely factors.

Are marketers faster with AI?#

Ahrefs reports that 85% of marketers using AI say they produce content faster: 47.9% say much quicker and 37.1% say slightly quicker. Just 2.5% say AI has slowed them down.

Ahrefs also reports the fastest growth among AI users in specialized tasks. Data analysis rose from 28% to 60%, image creation from 19% to 43% and keyword clustering from 23% to 45%. Ahrefs notes these were among the least popular uses in last year’s data.

Two of the contributors Ahrefs quotes disagree on whether speed is good. One says AI helps produce better content faster but still takes 4-6+ hours per piece. Another says faster means nothing without strategy behind it.

Which AI tools do they use?#

Among respondents who use AI, Ahrefs reports 83.2% use Claude, 75.7% use ChatGPT and 57.9% use Gemini. They are followed by Claude Cowork (42%), Claude Code (37%) and Codex (13%). The chart shows the six figures.

Ahrefs adds that 68% of AI users say they use skill files, and 55% say they create their own.

Ahrefs warns against reading this as market share. The 2025 survey asked which tool respondents used to create content, this year’s question allowed several answers, and some write-in tools were added to the totals.

How much do they trust AI content?#

Ahrefs reports that two-thirds (66%) of marketers trust AI-generated content only if a human reviews it first, and just 2.7% trust it with no review at all.

Among marketers who use AI, 6% say they generally distrust it, against 57% of those who do not use AI at all.

How much AI content is published?#

Across all respondents, 69% say they publish a larger share of AI content than a year ago. Among AI users the figure is 74%.

Ahrefs stresses that this does not mean more unedited writing: among respondents who say none of their published content is pure AI, 58% still report that their share has grown.

At the top end, 7% of AI users say more than 75% of what they publish is pure AI, without any editing or additional input, up from 4% in 2025. And 40% say none of their published content is pure AI, down from 50% last year. Ahrefs’ key takeaways describe the first group as content that is 75% AI or more, which is still the minority of what is published.

Does AI beat human writing?#

Ahrefs reports that 56% of AI users say human-written content is better quality than their AI content, and 23% say it is way better.

The answer depends on who is asked: 43% of founders and business owners say their AI content beats human writing, against 17% of junior and mid-level marketers.

Who checks and discloses AI use?#

Ahrefs reports that 29% (28.6% in its key takeaways) of marketers using AI now ask AI to check the quality of AI-generated content, up from 14% last year. The question allowed several answers. The most common check is still manual self-checking, and even fewer respondents said they do not check at all.

On disclosure, 78% of AI users do not tell website visitors that AI helped create their content, and 22% openly disclose it, up from 16% in 2025. Ahrefs says those publishing the most pure AI content were not any more likely to disclose it, and notes that the answer likely depends on what a respondent counts as AI content.

Is AI saving money on writers?#

Ahrefs reports that almost half of AI users say AI has not saved them any money on writers or freelancers, and 26% do not know whether it has. Its section heading gives the figure as 41% of respondents, so the base differs between the heading and the text.

Ahrefs explains that only 28% of AI users work with outside writers. Among those who reported savings, the typical saving is $1,001-$2,000 a month.

How reliable is the survey?#

These are the limits, partly stated by Ahrefs and partly our reading of the post.

• Small sample. There are 301 responses, and Ahrefs says last year’s survey had more, so it limits year over year comparisons to high-level ones.
• Self-reported. Every figure is what respondents said they do, not a measurement of what they publish. Speed, review habits and the share of AI content are all self-assessments.
• Unknown recruitment. The post does not say where the survey ran. A vendor’s survey tends to reach its own audience, which is our inference and not a statement from Ahrefs.
• Changed questions. Ahrefs notes that the tool question changed between years, and that the AI-checking question allowed several answers, so those answers can sum to more than 100%.
• Mixed bases. Some figures are shares of all respondents and others of AI users. The table above shows which.
• Chart-only data. We read the post’s text. Figures that appear only inside its charts are not restated here.

What should you take from it?#

This section is our own reading, not Ahrefs’ conclusion.

• Use it as direction, not as a benchmark. A sample of 301 whose recruitment is not described shows what a slice of marketers say they do. It does not give an industry rate.
• Treat review as the common practice. In this survey, 66% of marketers trust AI content only after a human reviews it. That matches Google’s 1 October 2026 wording that it is critical to manually fact-check and review AI-generated content, covered in our report.
• Watch the volume question. Using AI to produce many pages without adding value is a spam policy matter, discussed in our report on the spam update session. The survey says nothing about how any respondent’s pages performed in search.
• Check your own pages before you scale. Run drafts through the AI answer readiness checker to see whether they would be easy for an answer engine to quote. It scores structure only, so a human review still comes first.
• Read the wider topics. See content strategy and AI search for related coverage.