![]()
Here's a number that stopped me: 7%.
That's the share of marketers using AI who now say more than three-quarters of what they publish is pure AI content. No editing. No human fingerprints. Just hit generate and post.
A year ago that number was 4%. So it's not huge, but it's moving in one direction, and it's worth asking why.
Ahrefs, the SEO and AI marketing intelligence company, just released its second annual State of AI in Content Marketing survey. They polled about 300 marketing and SEO professionals in August 2026, including 68 in upper management, and the results landed on September 30, with a press release the next day. What they found is a strange mix of trust and suspicion happening at the same time.
More trust, more checking — at the same time
Here's the part that surprised me most. While that 7% is publishing raw AI output, 89% of AI users say they manually check everything AI produces. That's up from 80% last year. So confidence in AI is rising and the urge to double-check it is also rising. Both things are true at once.
Ninety-three percent of respondents said they use AI to make content now, up from 87% in 2025. And 24% say AI content is now better than human writing, up from 14% the year before. But 57% still say human writing wins. That's down from 65%.
Ryan Law, Ahrefs' Director of Content Marketing, put it simply: "Marketers are clearly becoming more confident in what AI can produce, but that doesn't mean they're handing over the keys."
I get that. I've used AI tools for research and brainstorming since I started writing more seriously about travel. It's fast. But fast isn't the same as right. Ahrefs' own content marketer, Louise Linehan, tried using AI to check AI-written work and came away less reassured, not more. "It just made me wonder what else AI had got wrong or missed," she said. That's a feeling I understand. The tool checking the tool doesn't exactly build confidence.
Who trusts it most? The people at the top
This is the part that makes sense once you think about it. Founders and business owners rate their AI content above human writing 43% of the time. Junior and mid-level marketers? Only 17%. The people furthest from the actual editing process trust the output the most.
Makes you wonder: who's doing the real quality control here?
Speed is the clearest win across the board. 85% of AI users say they're producing content faster, with almost half calling it "much quicker." Upper management reports even bigger gains, 62% say production sped up dramatically, and adoption in that group sits at 97%.
But here's a question nobody in the survey answers directly: does faster production without matching review time eventually create a gap? The data shows habits, not error rates. We don't actually know what's slipping through.
Readers barely know
The part that bugs me most is disclosure. Only 22% of AI users tell their audience that AI helped create the content, up from 16% last year. That means 78% say nothing.
Meanwhile, the EU's Article 50 transparency rules kicked in back in August 2026, with fines up to 15 million euros or 3% of global turnover for AI-generated text on public interest matters published without human editorial review. The survey never asked where respondents are based, so we don't know how many of them are even covered by that law.
As for tools, Claude leads the pack at 84% among AI users, ahead of ChatGPT at 75% and Gemini at 57%.
So what does this mean for you?
If you're reading content online right now, chances are good you've read something that was mostly or entirely AI-written, and you weren't told. That's not alarming on its own. Plenty of AI-assisted writing is solid. But it does mean the burden of figuring out what's trustworthy now falls more on you, the reader.
I've always believed in trying new tools, seeing what works, and adjusting. AI is no different. But the survey makes one thing clear: the people using these tools don't fully trust them either. That's probably the healthiest sign in the whole report.
Source: PPC Land
Still catching up? Start with Why Google Is Pushing Advertisers Toward Demand-Led Budgeting — And What It Means For the Rest of Us