- Is it worth publishing AI content without human review?
- What the Ahrefs research really says: the numbers that matter
- What does your company actually risk if you don't review AI content?
- How does human review that truly protects you work in practice?
- Three concrete scenarios: how does the situation change for an Italian company?
- Which tools help you manage AI content review?
- Why are people talking about this now and not a year from now?
- How to build a review process: the concrete steps to get started
- What are the most common mistakes in managing AI content?
- How do you know if your review process is working?
- So, is it worth setting up a review process for AI content?
- How we see it at SHM
- Frequently Asked Questions
- 93% of marketers already use AI to create content, but 89% keep a human review in place: oversight is still the norm, not the exception.
- Publishing AI text without editing exposes you to three concrete risks: lost rankings on Google, fines of up to 15 million euros under the European AI Act, and a drop in audience trust.
- Building a structured review process — not just a quick reread — is now the most important skill for anyone using AI in marketing.
- 76% of Italian SMEs have not yet invested in AI: those who start now can do it with the right rules from day one.
7% of marketers who use AI publish text exactly as it comes out of the tool, without touching it. A year ago it was 4%. If your company produces content with AI, this trend affects you directly — even if you're not part of that 7%.
Is it worth publishing AI content without human review?
No, almost never. The risks — to reputation, rankings and legal compliance — far outweigh the time saved. Human review remains the main quality control tool, even among people who use AI every day.
Ahrefs' annual survey of 300 marketing professionals, published on September 30, 2026, paints a clear picture: 93% of marketers use AI to create content, but 89% manually check what the AI produces before publishing it. Only 7% publish AI text without significant changes.
That 7% is up from 4% in 2025. It's not the majority yet, but the direction is clear. And the consequences of that choice — for your brand, your rankings and compliance with the new European rules — are real.
| Your situation | Is it worth publishing without review? | why |
|---|---|---|
| You publish content on technical or industry topics | No, for now | AI gets details wrong: a technical error damages credibility more than a publishing delay |
| You produce corporate or brand content | No, for now | Brand tone and voice can't be replicated without human input |
| You use AI for internal drafts or brainstorming | Yes | For internal use, formal review isn't necessary |
| You publish on matters of public interest (health, finance, law) | No, for now | The European AI Act requires labeling or substantial editorial review |
| You have a dedicated editor who reviews every piece | Yes | With structured review, AI speeds things up without lowering quality |
What the Ahrefs research really says: the numbers that matter
The most significant finding isn't the 7% who don't review, but the fact that 89% keep human review even as trust in AI grows. Speed goes up, oversight doesn't go down.
Ahrefs surveyed 300 marketing and SEO (search engine optimization) professionals in August 2026. The results show a market that uses AI more and more intensively but hasn't given up human control. You can read the source's full analysis on PPC Land.
One figure deserves attention: 29% of marketers use one AI to evaluate text produced by another AI — a practice that has doubled from 14% in 2025. The problem is that this approach doesn't replace human review. As an Ahrefs content marketer who tested the method observed: even when the AI flagged problems, the feeling was of not knowing what else it had gotten wrong or missed.
Speed is the clearest advantage: 85% of marketers produce content faster, and among executives the share rises to 62% saying they work “much faster”. But speed without strategy, as an Ahrefs SEO educator put it, “smells like a factory”.
What does your company actually risk if you don't review AI content?
Three distinct, measurable risks: a drop in Google rankings, legal fines of up to 15 million euros, and a loss of audience trust that a recent study found doubled in just one year.
The risks aren't theoretical. They fall into three separate areas, each with concrete consequences for your business.
| Risk area | What happens | Extent of the damage |
|---|---|---|
| Google rankings | Google's Search Quality Rater Guidelines (updated January 2025) give the 'Lowest' rating, the worst score, to pages whose main content is auto-generated with no original added value | Loss of organic visibility, traffic and leads |
| EU AI Act | From August 2, 2026, anyone publishing AI-generated text on matters of public interest without substantial editorial review must label it as AI-generated | Fines of up to €15 million or 3% of global turnover |
| Brand trust | According to a Fractl study (2026, 1,008 US consumers), the share of people who say heavy use of AI would reduce their trust in a favorite brand has doubled: from 20% in 2025 to 40% in 2026 | Eroded reputation, higher churn |
On the legal side, Article 50 of the AI Act (the EU Regulation on artificial intelligence) includes an important exemption: if AI text has undergone substantial human review by someone who takes editorial responsibility, the labeling requirement doesn't apply. The review must include fact-checking, not just spell-checking. You can read the details of the European Commission's guidance on Altalex and the operational deadlines on CityNext.
To dig deeper into how Google evaluates AI content, the analysis of the Quality Rater Guidelines is available on Fullstackoptimization.
How does human review that truly protects you work in practice?
Proofreading isn't enough. The review that counts is the one that adds value: fact-checking, brand voice, original examples. That's what satisfies both Google and the AI Act.
There's a clear difference between a formal review and a substantial one. The first is a spelling and readability check. The second turns AI text into content that carries the company's editorial signature.
- 1 · Internal writer or expertFact-checkingChecks every figure, statistic and technical claim. AI often invents numbers that sound plausible but are false (so-called 'hallucinations').
- 2 · Writer or marketing managerBrand voice adaptationRewrites the parts that sound generic. Adds concrete examples, anecdotes or references to the Italian market.
- 3 · Subject-matter expert (if available)Technical reviewFor specialist content (legal, medical, financial, technical), an expert verifies substantive accuracy.
- 4 · Editor or content leadEditorial approvalWhoever signs off on the content takes editorial responsibility. This step is what triggers the exemption from the AI Act labeling requirement.
- 5 · SEO or marketing teamOptimization and publishingAdds meta tags, internal links and optimized images. Only then does the content go live.
Three concrete scenarios: how does the situation change for an Italian company?
The time savings are real, but they depend on how you structure the process. Without review, those savings turn into a hidden cost.
Illustrative examples: numbers are meant to show the reasoning, they are not real customer data.
| Company type | Current use of AI | Risk without review | Recommended approach |
|---|---|---|---|
| Fashion e-commerce (20 employees) | Product pages and category descriptions generated with AI | Duplicate or generic content penalized by Google; loss of organic traffic | Quick review (10 min per piece) to add original details and brand voice |
| Professional firm (accountant, lawyer) | Blog articles on tax or legal regulations | Factual errors on regulated topics; high reputational and legal risk | Mandatory review by an expert professional before every publication |
| B2B manufacturing company | Data sheets, case studies, trade show content | Wrong technical data that undermines credibility with buyers | Technical review by the engineering department + marketing approval |
Which tools help you manage AI content review?
There are tools for every stage of the process: from generation to verification, all the way to editorial management. The right choice depends on your content volume and team structure.
| Phase | Tool / approach | What it does | Limit to keep in mind |
|---|---|---|---|
| Generation | ChatGPT, Claude, Gemini, Ahrefs Letaido | Produces drafts, outlines, text variations | Doesn't verify facts: the output must always be checked |
| Fact-checking | Manual research + primary sources | Checks every data point cited by the AI | No automatic tool is 100% reliable |
| Plagiarism/originality check | Copyscape, Originality.ai | Checks that the text isn't copied from other sources | Doesn't detect factual errors or hallucinations |
| AI reviewing AI | Review features built into AI tools | Flags style or consistency issues | Doesn't replace human review; can give a false sense of security |
| Editorial management | Notion, Trello, Asana with approval workflows | Tracks who reviewed and approved each piece of content | Needs team discipline to be effective |
Why are people talking about this now and not a year from now?
Three factors are converging in 2026: AI adoption in marketing teams has exploded, European rules have come into force, and consumer trust in AI content is falling. Those who wait lose ground on all three fronts.
The Italian artificial intelligence market reached €1.8 billion in 2025, up 50% on the previous year, according to the Artificial Intelligence Observatory at Politecnico di Milano. Generative AI, the kind that produces text, images and video, accounts for 46% of the total.
Yet according to the Digital Innovation in SMEs Observatory at Politecnico di Milano, 76% of Italian SMEs have neither invested in AI nor plan to, and only 7% have launched structured training programs. If you start now, you can build the right process from day one, without having to unlearn bad habits.
At the European level, according to the IAB Europe report on the impact of AI on digital advertising (2026), 86% of organizations already use AI for marketing and 78% have appointed a formal AI governance lead, meaning someone who oversees how AI is used and controlled. In Italy, this role is still rare in SMEs.
How to build a review process: the concrete steps to get started
You start with an audit of what you already publish, define who reviews what, and build a workflow that scales with your content volume.
AI content governance ties into the broader topic of editorial strategy. If you're considering how to position your content in AI engines too, you can explore the topic on Marketing e performance ADV, our cluster dedicated to paid media, content and performance.
- 1 · Marketing manager or content leadAudit your existing contentList every channel where you publish content (blog, social, product pages, newsletters). Identify which ones already use AI and what process they follow.
- 2 · Marketing teamDefine risk levelsSort your content by risk: high (technical, legal and health topics), medium (brand content), low (internal use, brainstorming).
- 3 · Content lead + legal (if needed)Write your internal guidelinesDecide who can use AI, for which content, and who must approve before publication. One page is enough to get started.
- 4 · Marketing teamChoose your tools and workflowUse a project management tool (even just a shared spreadsheet) to track the review status of every piece of content.
- 5 · The whole teamTrain your team on the processEven one hour of internal training reduces mistakes. The 7% who publish without review often do it because there is no process, not by deliberate choice.
- 6 · Marketing managerMonitor and updateReview your guidelines every quarter. AI evolves fast: what works today may not be enough six months from now.
What are the most common mistakes in managing AI content?
The most common mistake isn't using AI, but using it without a defined process. The second is believing that AI reviewing AI is enough.
The question of trust in AI content also emerged in our analysis of creator and AI content: more than half of consumers prefer it — a sign that audiences can already tell authentic content from generic content. And if you're evaluating how AI changes your relationship with your audience on the advertising side too, it's worth reading our analysis of ChatGPT Ads click-to-chat and conversion without website traffic.
| Common mistake | Why it is a problem | How to avoid it |
|---|---|---|
| Publishing without a defined process | Each team member decides on their own how much to review, leading to inconsistent results | Write internal guidelines, even short ones |
| Using AI to review AI | AI doesn't detect its own hallucinations; it gives a false sense of security | The final review must always be human |
| Treating all content the same way | A social post and a technical white paper carry very different risks | Sort content by risk level and adapt the process |
| Not updating the guidelines | AI capabilities change quickly; yesterday's rules may not be enough | Quarterly review of internal guidelines |
| Ignoring the AI Act labeling requirement | It has been in force since August 2, 2026: not knowing the rule doesn't exempt you from responsibility | Check whether your content falls within the scope of Art. 50 |
| Not training the team | The 7% who publish without review often do it out of habit, not by choice | Even a one-hour training session changes behavior |
How do you know if your review process is working?
Three indicators are enough to monitor the health of your process: perceived content quality, organic ranking and the rate of errors caught in review.
| Indicator | How to measure it | Red flag |
|---|---|---|
| Rate of errors caught in review | Count the factual errors found before publication each month | If it rises, AI is producing less reliable content or the briefing is inadequate |
| Organic ranking of AI-assisted content | Track the Google positions of content produced with AI | A systematic drop signals quality problems as perceived by Google |
| Average review time per piece of content | Track how long each reviewer takes | If it exceeds the writing time, the briefing to the AI needs improving |
| Audience feedback | Comments, shares, requests for clarification | An increase in questions about technical details signals superficial content |
| AI Act compliance | Make sure every piece of content on topics of public interest has documented review | Any content published without a review trail is a legal risk |
So, is it worth setting up a review process for AI content?
Yes, for almost every company that already uses AI for content. The question isn't whether to do it, but how structured it should be based on volume and risk.
It's worth it if you already use AI to produce content published on your website, blog or social channels, and you don't yet have a formal review process. The legal and reputational risk justifies the investment even for small teams.
A middle ground: you produce AI content only for internal use or for low-risk channels (informal social, short newsletters). A lightweight process, even just a checklist, is enough for now.
Not yet if you don't use AI for content yet, or you only use it for brainstorming and outlines that are never published directly. You can wait, but it pays to prepare before you scale.
How we see it at SHM
For us, AI is a tool that gives time back, not one that replaces judgment. We believe the value of content doesn't lie in how fast it was produced, but in its ability to answer a real question from the reader with precision and authority. That doesn't change with AI: the only thing that changes is who does the routine work and who does the work that matters.
When we work with companies that want to bring AI into their editorial processes, the first step isn't choosing the tool: it's understanding which content truly needs a human voice and which can be delegated. Think of a B2B manufacturing company that produces data sheets in five languages: AI can generate the structure and the first draft, but an engineer's technical review and editorial localization (not mere translation) are the value that no tool can replicate automatically.
If you want to understand how to structure this process in your company, you can start with a consultation on strategic AI integration, or explore how to optimize your content for the new AI engines with our SEO for AI service. For those who want to train their internal team, the AI training for companies program is the most practical starting point.
AI content governance also overlaps with visibility: if your content isn't being cited by AI engines, it's worth reading our analysis of ChatGPT Ads in Asia and how to assess it for your strategy, and of Amazon DSP on ChatGPT Ads and the opportunities for Italian B2B.
Frequently Asked Questions
It depends on the type of content and how much you've reviewed it. From August 2, 2026, the European AI Act (Art. 50) requires AI-generated texts on topics of public interest to be labeled if they haven't undergone substantial editorial review. If you have an editor who checks the facts and takes responsibility for the content, the labeling requirement doesn't apply. For standard commercial content (product pages, social posts), the rules are less strict, but transparency is still recommended.
Google doesn't penalize AI as such, but it does penalize low-quality content, regardless of how it was produced. Google's Quality Rater Guidelines (updated January 2025) rate pages with automatically generated content that adds no original value as 'Lowest'. A reviewed AI text, enriched with real examples and fact-checked, is treated like any other quality content.
It depends on how complex the content is. For an 800-word blog post on a non-technical topic, a substantial review takes 20-30 minutes on average: fact-checking, adjusting the tone, adding original examples. For technical or legal content, the time can double. Either way, it's far less than writing the text from scratch.
No, and the data confirms it. 29% of marketers already use this approach, but those who have tested it report that AI doesn't detect its own hallucinations (that is, information that is made up but plausible). The final review must always be human, at least for fact-checking. AI can help check style and consistency, but it doesn't replace editorial judgment.
Internal-use content — brainstorming, outlines, first drafts, document summaries — can be handled with AI without formal review. For published content, zero risk doesn't exist, but it is very low for short, low-impact formats (e.g., social captions on non-controversial topics). Risk grows with technical complexity, legal relevance and the visibility of the content.
AI Act Art. 50 applies to AI-generated texts on 'topics of public interest' — a category that includes health, finance, law and politics, but may extend to other sectors depending on interpretation. If you publish content on regulations, financial products, health or safety, it's wise to consult a lawyer. For all other commercial content, substantial editorial review is still the safest choice.
Human review isn't an extra cost on top of AI. It's the step that turns generic copy into content that carries your company's name.
Related articles
Discover more articles exploring similar topics, selected to offer you a more complete and stimulating perspective. Each piece of content is carefully chosen to enrich your experience.
