When the algorithm devours the clicks – the real price of AI Slop

What threat does AI Slop pose to content creation – and why is authenticity the only weapon that the algorithm cannot replicate?

Perhaps the internet people do not yet feel Will Smith as so old. of the spaghetti-eating video. grotesque imagery, in which The Exorcist c. film's scenes can be seen to shameful motion coordination of a supposedly human figure. In contrast, by 2026, the suddenly touching AI-generated video stories and AI influencers have flooded the internet. Alongside this, the beauty industry has also broken into the market with highly realistic image quality, capable of imitating detailed skin textures, for example, in printed media featuring face and hand models. Therefore, we can rightly ask, what is the probability that what is presented to our eyes when opening our favorite social apps or even the news is real?

The dead internet is no longer a theory – but statistics.

Let's look at the current numbers. In the first quarter of 2026, the Graphite SEO company examined 55,400 URL articles during its research. To minimize the margin of error and false tagging,hamis címkézésthey performed cross-validation with 3 different AI detectors for maximum accuracy. According to the results, 49.91% of the textual content found on the internet is already AI-generated. Robots write, edit, and read our content and rank it day by day. Every human being using the internet can rightly ask whether the dead internet theory is more than just a theory.

A highly debated factor is what plays a more decisive role in terms of content and website optimization when focusing on SEO or GEO, and how unfairly AI relates to human-generated posts. But what drives this machinery? Recommendation algorithms no longer ask: is this content useful to us or who wrote it? Instead, they ask whether it satisfies the search intent and stops scrolling for 1.2 seconds.

If an AI tool can generate a thousand SEO articles or posts in a single afternoon, the content becomes a mass commodity in the market.

The bland content, clichéd phrases, and tasteless unstructured sentence structures all point in one direction in content creation.

This phenomenon is referred to in common language, and in 2025, it was also recognized by the Merriam-Webster dictionary 12. (Great Divide): The phenomenon where search volume increases while the number of clicks decreases. Users are searching more than ever but leaving Google pages less frequently.as the word of the year. However, raw, unedited AI content is unsustainable in the long run. There is a consensus among professionals that mass-generated texts and automated language patterns become almost immediately devalued after Google updates, or are not indexed at all, as they do not provide so-called informational surplus. (Információs nyereség).

The new golden rule of content production could become the so-called “Sentence test”. According to this, if any sentence of your article can appear on your competitor's page with the same wording, the content is not differentiated enough to rank higher. Although there are large brands in the market that can succeed with raw AI content due to their domain strength and backlinks, this is only a short-term success. The sustainable solution may be offered by the hybrid model. Accordingly, artificial intelligence should be exclusively entrusted with research and drafting tasks (along with the appropriate model selection). It becomes essential for the final form to be shaped by a flesh-and-blood editor who adds the final touches such as emotional charge, logic, and unique perspective.

But where have the clicks gone? - The black box paradox.

The problem may not only be the oversupply of AI slop but also the invisibility of higher-quality content. There is a phenomenon referred to in the literature as” A Nagy Leválás”which essentially means: search volume is increasing, but the number of clicks is decreasing. Users are searching more than ever, but they are leaving Google’s page less and less. The reason for this phenomenon is that in AI-driven search results, they receive answers before reaching the page, thus “devouring” the clicks.

Looking at the statistical data, by 2026, traditional Google searches will end without clicks 60% of the time. Where AI Overview appears on the results page, this can also mean 83%. For searches initiated in Google AI Mode, this ratio is 93%. According to Pew Research, users click on links in the AI Overview only 1% of the time. One reason for this is early information dissemination, and the other is blind trust in the accuracy of the results. However, it is worth approaching the latter critically due to the occurrence of bad linking and hallucinations. The generated summaries may contain false information 9-10% of the time, and source citations are incorrect or irrelevant in more than 56% of cases.

Just think about it: an article that provided the backbone of organic traffic for years now sits at the top of the search results, and 83% of users never click on it because Google has already answered their question on the results page. According to Ahrefs data, the informative nature (Informational intent.In the case of content, the AI Overview has reduced clicks to the 1st place page by 58% by December 2025, which is a huge drop compared to the 34.5% of the spring position in 2025. The decline in the display rate and click-through rate has also taken on different sizes by industry. In the healthcare sector, a significant drop of up to 80% has been observed in the education sector at 43%.

In e-commerce or purchase intent queries, the AI Overview's rate is surprisingly only 3.2%. In these searches, Google has regained its leading role, as users want to click and reach the product page, not just read.

We can state that the use of AI summaries significantly affects the CTR for both paid and organic results. The percentage of users who ultimately reach our page depends on the combination of the search intent-target entered into the AI summary and the quality of the references.

GEO or SEO?

What can we do if the black box swallows the content? Let's optimize for it!

Besides the importance of traditional SEO, the goal is no longer just to appear on the first page of our website, but for the AI to cite it.

Brands that are marked as reference sources by AI Overviews receive 35% more organic and 91% more paid clicks than those that are not cited. In GEO, compared to SEO, we need to influence the selection mechanisms of AI systems. Brands must clearly answer important questions such as what value they represent and what solution they provide. Vague positioning reduces the chances of the model referencing the company. This is particularly critical for franchises or fragmented brands, where the corporate website, local subpages, and PR appearances must communicate uniformly so that the AI knows what to accept as a credible source.

Furthermore, the role of earned media is becoming more valuable. How naturally the brand appears organically on online forums (e.g., Reddit) and review sites (e.g., G2) and industry publications.

Structured content and schema markup are given a prominent role.Schema MarkupAdding schema markup to websites makes the content eligible for search engines like Google to display expanded search results, also known as rich snippets(also known as rich results). Examples include star ratings, prices, and whether the product is in stock.

This also shows that AI models prefer to cite content that is clearly structured, thematically deep, and supported by external sources when discussing a topic.

GEO cannot yet be treated (and is not worth treating) as a separate item from SEO today. Since it functions as an extension of our existing strategy rather than a new channel. As user search habits evolve with new momentum, we must also adapt. In addition to classic KPIs, new success metrics include: prompt coverage (how many relevant questions we appear in the answers) and the recommendation rate (does the system check if the user is looking for a solution with a specific purchase intent) will also be included in the appropriate column of the dashboard.

Generation Z and credibility against Faux media.

1. Returning to the spaghetti-eating Will Smith. In videos of this kind, we laughed or deemed them a failure in 2023. Then in 2025, the Google Veo version 3 seemed artificially perfect. Now in 2026, the Kling 3.0 can generate a complete scene from a single prompt with seamless dialogue, natural facial expressions, and movement.

2. The 2025 Frontiers research, involving 104 participants and using 50 images, showed that people recognize AI-generated images on average at 63.7%. But for the best models, this ratio can drop to 29%. A more shocking statistic is that 71% of participants considered the most advanced AI images to be real, as if the boundary between real and fake visuals had blurred. However, this does not mean that the audience, especially younger generations, blindly accepts everything presented to them. The current situation is more complex than that.

3. Generation Z (1997-2012) already uses AI tools at 70%. In contrast, consumer demands are changing. According to InMobi's entertainment industry study, 61% of Generation Z consumers expressed a strong and definite preference for user-generated content (UGC).4. Furthermore, 56% of this age group specifically seeks to buy from brands that transparently label their media and prioritize human writers and designers. Additionally, 72% are particularly distrustful of AI security protocols, making them extremely defensive against automated data tracking. 5. By 2026, the audience has reached a quieter resistance and the appeal of imperfection. This is the true paradox that visual marketing faces, as the "flawless" image becomes suspicious. Poreless skin, symmetrical faces, and flawless settings do not convey reliability and professionalism but rather distance. Gen Z prefers platforms (BeReal, direct story formats, "direct blunders" in videos) that tell stories of awkwardness, human nature, and captured moments.

6. Summary – therefore, a lot of “good” content is not enough.” “direct blunders” in the videos), which narrate awkwardness, human nature, and the moment captured in time.

8. Instead of saying goodbye, AI seems to remain in our daily lives, whether at work, entertainment, or administration. The real question today is not how a brand can create the most or the most perfect content or posts the fastest. But how authentic, human, and reliable it remains for its consumers. The clearest sound breaking through the online static noise is still human.

Today, as consumers and brands, we live in the grip of algorithms. We simultaneously struggle with the overwhelming surplus of information, making our content visible against visual noise and distrust.

10. AI Slop: Weak – low-quality, unwanted, artificially generated content.

11. (Merriam-Webster, Word of 2025)

12. (Great Divide): The phenomenon where search volume increases while the number of clicks decreases. Users are searching more than ever but leaving Google pages less frequently.13. AI Overview: A summary response generated by artificial intelligence that appears on Google's search results page, answering the search query directly on the results page. 14. Black box: A metaphor for how Google Gemini does not rank but absorbs visitors and clicks. The content appears on the results list but remains invisible because the user never reaches our page.
A Nagy Leválás 15. GEO.
AI Overview: A summary answer generated by artificial intelligence that appears on a Google search results page, answering the search query directly on the results page.
Black box: a personal metaphor that Google Gemini does not rank but absorbs visitors and clicks. The content appears in the results list but remains invisible because the user never reaches our page.
GEO (Generative Engine Optimization): The next stage in the evolution of SEO. Its essence is not ranking, but citation: the goal is not for the user to click through to the website, but for the AI response to reference it.
Schema Markup (Schema Markup): A structured data markup given to a webpage that allows search engines to interpret and display the content in an enhanced format on the results page.
Rich Snippets (Enhanced results): Enriched search results that appear as a result of schema markups, such as star ratings, prices, stock information, which convey more information directly on the results page than the traditional blue link.
Information Gain (Information Gain): The added value that a piece of content contributes compared to existing internet information. In their absence, Google's algorithms do not index or downgrade the given content.
UGC (User Generated Content): Content created not by the brand, but by its audience, such as reviews, photos, videos, and ratings.