Facebook Pixel
(303) 578-6256

What Percentage of Online Content Is AI Generated in 2025?

AI Content

The Truth Behind a Very Different Internet

If the internet has started to feel repetitive, strangely familiar, or suspiciously robotic, your instincts are working. In 2025, the web is overflowing with AI generated writing, videos, images, profiles, and entire networks of bots pretending to be people. The numbers are big, the trend is accelerating, and the consequences reach far beyond SEO chatter.

This guide breaks down what we actually know, using real data from crawls, platform reports, academic studies, and industry analysis. We will look at AI written articles, synthetic video, fake profiles, bot traffic, and what all this means for businesses trying to operate in a world where machines are producing a huge share of what humans consume.

Let’s get straight to the numbers.

1. So what percentage of online content is AI generated in 2025?

There is no single master statistic, but the most reliable studies give us a clear picture.

AI in newly created web pages

Ahrefs analyzed nearly a million new web pages published in April 2025 and found that 74.2 percent contained detectable AI generated content.
Source: https://ahrefs.com/blog/ai-content-prevalence

AI in newly published articles

Graphite, an SEO firm, examined more than 60,000 new articles from 2020 through 2025 and found that by late 2024, more than half of new English language articles were primarily AI written. That percentage grew into 2025.
Source: https://graphite.io/blog/ai-content-study

AI in all existing web text

One widely cited analysis estimates that about 57 percent of all online text has been generated or translated using AI tools. This includes machine translation, rewriting, and generative writing.
Source: https://news.ycombinator.com/item?id=40384241

Future projection

Europol and other analysts have warned that up to 90 percent of online content may be synthetically generated by 2026.
Source: https://www.europol.europa.eu

The simple 2025 takeaway

AI dominates new content creation. Humans still dominate the historical archive of the internet, but that balance is flipping fast.

2. Why these numbers are hard to measure accurately

Before anyone uses these stats to panic at conferences, there are caveats worth understanding.

Detectors are not perfect

AI detection tools work by pattern matching. They misclassify edited AI text as human writing and formulaic human writing as AI.

AI involvement does not always mean AI authorship

Many pages are human drafted but AI translated, AI summarized, or AI polished.

New content is not the whole internet

Millions of old blog posts, forums, PDFs, and scanned books dilute AI’s total share when you look at the entire web.

Platforms have incentives to downplay AI usage

Independent research tends to paint a clearer picture than corporate PR.

Even with these limitations, the trend is obvious. Machines are doing more and more of the creating.

3. AI written SEO content: the biggest and fastest growing category

This is where AI has taken over the fastest.

  • More than 74 percent of new pages contain AI writing.
  • More than half of new articles are primarily AI generated.

But here is an important twist that human readers should care about:

AI written content rarely ranks well in Google

Graphite’s study found that while AI generated articles are published in huge quantities, top ranking pages in search results still tend to be human written or heavily human edited.

In other words, the web is filling up with AI text, but quality and originality still matter more than volume.

If you run a business, this is great news. You do not have to out spam robots. You have to outperform them.

4. AI generated video: YouTube, TikTok, Shorts and the deepfake boom

Video is harder to quantify, but reliable sources give us real signals.

AI in YouTube Shorts

A recent industry survey found that about 42 percent of Shorts creators use AI tools for editing, captioning, enhancement, or full video generation.
Source: https://influencermarketinghub.com/youtube-shorts-statistics

This does not mean nearly half of Shorts are fully AI created, but it shows how deeply AI is integrated into production.

AI driven YouTube “slop channels”

Reporting has revealed that around 10 percent of the fastest growing channels rely heavily on AI generated scripts and visuals.

TikTok and synthetic media labeling

TikTok has labeled more than 1.3 billion AI generated videos, and is now testing tools that let users reduce AI content in their feed.
Source: https://www.theverge.com/2024/2/28/24146513/tiktok-ai-generated-content-labeling

Deepfakes

Synthetic videos are now mainstream.

  • Over 500,000 deepfake videos were shared online in 2024.
  • Platforms like YouTube have removed entire clusters of deepfake Bollywood videos that reached tens of millions of views.

The trend is crystal clear. AI video creation is exploding, and major platforms are scrambling to keep up.

5. Fake social media profiles and AI generated faces

This is where AI and deception intersect.

How many social media accounts are fake?

Depending on the platform, estimates vary:

  • Facebook reports 3 to 5 percent of monthly active users are fake.
  • Meta removes hundreds of millions to billions of fake accounts every quarter.
  • Elon Musk claimed 20 percent of Twitter accounts were fake during acquisition negotiations, although this number was disputed.
  • Independent researchers warn that the true share could be higher because platform access for auditing is limited.

AI generated profile photos

Studies analyzing millions of Twitter profile images found:

  • Around 0.02 to 0.05 percent of accounts use AI generated faces as profile pictures.
  • That sounds tiny, until you remember scale. A fraction of a percent on platforms with billions of users still translates to millions of AI created personas.

Public perception

Survey research shows 80 percent of people report encountering suspicious or fake profiles regularly.

This environment is fertile ground for scams, impersonation, and misinformation.

6. Bot traffic: machines reading machine generated content

AI generated content is only half the story. A huge portion of traffic is also automated.

Industry reports show:

  • Roughly half of all internet traffic is now non human.
  • Around one third of all traffic comes from malicious or unwanted bots.
  • AI models have made it easier than ever to create large, sophisticated bot networks.

We are not just dealing with an internet full of AI produced content. We are also dealing with an internet where bots consume and interact with that content.

This is how you end up with machines talking to machines while people try to make sense of the noise.

7. What all this means for businesses and real humans

There are a few unavoidable implications.

1. Volume is meaningless now

If someone used to brag about publishing 40 blog posts a month, that number meant something. In 2025, it signals absolutely nothing. Anyone can dump hundreds of AI generated pieces online.

2. Trust becomes your competitive advantage

In a world full of synthetic content, authentic voices stand out.

3. Security risks grow with synthetic content

Fake accounts, synthetic identities, deepfake scams, automated phishing, impersonation. All of these are growing because AI lowers the cost of deception.

4. Google is already rewarding the opposite of AI slop

Human expertise, originality, firsthand experience, and well sourced data are ranking signals that AI systems cannot cheaply mimic.

5. Human judgment is the differentiator

AI can generate words, images, and videos, but it cannot replicate lived experience, strategic thinking, or ethical insight. Businesses that blend AI efficiency with real expertise will win.

8. How organizations can stay sane in an AI heavy internet

Here are practical, no nonsense steps for operating in this new landscape.

Audit your content workflows

Know where AI is being used and create clear quality and disclosure standards.

Publish content that contains experience, not just information

AI can only remix what already exists. It cannot generate your case studies, stories, insights, or customer results.

Protect your online presence

Implement bot detection, rate limiting, two factor authentication, and stronger internal verification protocols.

Train your staff on synthetic media awareness

Most companies are still unprepared for fake calls, cloned voices, synthetic profiles, or AI generated impersonation.

Use AI carefully, not blindly

AI is a powerful tool for drafting, summarizing, and analysis, but it needs human review for accuracy, ethics, and brand context.

FAQ: AI generated content and online authenticity (optimized for quick answers)

Q1. How can I tell if content was written by AI?
There is no perfect detector. AI detection tools provide probability scores, but the safest way to judge content is by checking for author transparency, verifiable sources, and depth of insight.

Q2. Do AI generated articles rank well on Google?
Not usually. Large scale studies show that although AI articles are published in huge volumes, the top ranking pages tend to be human written or heavily human edited because they offer more originality and value.

Q3. Are deepfakes a major problem today or only a future threat?
They are a current and fast growing problem. Hundreds of thousands of synthetic videos circulate online, and many are used in scams, harassment, or misinformation.

Q4. Are fake social media profiles mostly bots or humans?
Both exist, but AI is accelerating the creation of automated profiles. Some use AI faces, AI writing, and AI behavior modeling to appear more realistic.

Q5. Will human creators still matter if AI generates most online content?
Yes. AI can produce volume, but humans provide originality, lived experience, strategy, ethics, and narrative voice, all of which remain essential.

This article was prepared by ITGuys IT Support and Consulting, a Denver based provider of managed IT services and technology strategy for small and midsize organizations.