More articles are now created by AI than humans

More articles are now created by AI than humans

Since ChatGPT launched in November 2022, many companies have explored publishing content generated by LLMs such as ChatGPT, Claude, and Gemini to grow their traffic across channels such as Google Search, social, and advertising. This is a cost-effective alternative to spending hundreds of dollars for humans to write content.

The quality of AI content is rapidly improving. In many cases, AI-generated content is as good or better than content written by humans, according to an MIT study. It is often hard for people to distinguish whether content is created by AI, per findings from an Originality AI study.

A research team from Vertical AI growth agency Graphite evaluated the prevalence of AI-generated articles from a dataset spanning January 2020 until May 2025. Their research reveals that in November 2024, the quantity of AI-generated articles being published on the web surpassed the quantity of human-written articles.

The research indicates significant growth in AI-generated articles coinciding with the launch of ChatGPT in November 2022. After only 12 months, AI-generated articles accounted for nearly half (39%) of articles published, per one Graphite evaluation.

AI-Generated Article Growth has Plateaued

While AI-generated articles grew dramatically after ChatGPT launched, Graphite’s research suggests that trend will not continue. Instead, the proportion of AI-generated articles has remained relatively stable over the last 12 months. This may be because practitioners found that AI-generated articles do not perform well in search, as shown in a separate study.

The Methodology Behind the Findings

In order to compile the research, the team needed a representative sample of English-language articles on the web. To do so, they randomly select 65,000 URLs from CommonCrawl, one of the largest publicly available web archives. They confirmed that each URL is in English, has an article schema markup, is at least 100 words, has a publish date between January 2020 and May 2025, and is an article or listicle (as classified by Graphite’s page type classifier).

AI Detection Algorithm

Accurate detection of AI-generated content is required to make claims about the prevalence of AI-generated articles on the web. There is a considerable disagreement about the accuracy of AI detection algorithms, and many argue that detecting AI is impossible, or at best, highly inaccurate. Many companies offer AI detection algorithms, including Originality.ai, GPTZero, Grammarly, and Surfer.

To compute the percentage of AI-generated content in an article, the research team used the same algorithm described in their 2024 whitepaper, but classify each chunk using Surfer’s AI detector with a chunk size of 500 words. The researchers classified an article as AI-generated if the algorithm predicts that more than 50% of the content is AI-generated, and human-written otherwise.

Evaluation of False Positive Rate

More from this section

Woman rescued from Alaska home following devastating typhoon

+3

Teens who don’t sleep enough at greater risk of suicide: study

The early signs of frontotemporal dementia

Before classifying the articles in the dataset, the research team evaluated the accuracy of Surfer’s AI detection algorithm.

To evaluate the false positive rate (the percentage of human-written articles classified as AI-generated), Graphite leveraged a dataset of human-written articles. Since the large-scale adoption of AI tools began with ChatGPT, the researchers made the assumption that articles published before the release of ChatGPT had a high probability of being written by humans. Therefore, the researchers ran Surfer’s AI detector on the 15,894 articles in Graphite’s CommonCrawl dataset that were published between January 2020 and November 2022. SurferSEO’s AI detection tool classifies 4.2% of these articles as primarily AI-generated, suggesting a 4.2% false positive rate.

Evaluation of False Negative Rate

To evaluate the false negative rate (percentage of AI-generated articles classified as human-written), the team used OpenAI’s GPT-4o to generate 6,009 articles on a wide range of topics, including commerce, finance, consumer, and B2B enterprise.

The researchers used the OpenAI API to generate the articles using the system prompt:

“You are an expert content writer. Your task is to generate clear, engaging, and informative content about the topic provided by the user. Write in a professional yet friendly tone. The target audience is people searching on the web for key terms related to the topic provided by the user. The user will provide a word count for the prompt. Ensure that the generated content adheres to the specified word count, allowing for a variance of plus or minus 10 percent. Avoid jargon unless explained. Do not include any disclaimers or meta-commentary.”

The team then prompted, “Write an article on the topic ‘{topic}’ with approximately {word_count} words.”, with word_count set to the number of words in a reference human-written article on the same topic.

Based on the data collected, SurferSEO’s AI detection algorithm correctly classifies 99.4% of the AI-generated articles as AI-generated, suggesting a 0.6% false negative rate for GPT-4o.

Quantifying AI-Generated Articles on the Web

Finally, the team classified all 65,000 articles in the dataset to evaluate the percentage of articles being published on the web that are AI-generated.

Limitations to Be Considered

Many people incorporate AI into their content creation process. One strategy is to ask AI to create a first draft, then have a human in the loop to edit or rewrite it. The research team study did not evaluate the prevalence of content created using this strategy, and AI-generated human-edited articles may be even more prevalent.

Additionally, AI models continue to improve, and may become more difficult to detect. The Graphite team only evaluated the false negative rate on articles generated by GPT-4o. The AI detection algorithm may have lower accuracy when applied to articles generated by other models.

This story was produced by Graphite and reviewed and distributed by Stacker.