There is no single audited global number for “AI news” per month, and pretending otherwise would be sloppy. Different trackers count different things: some count articles whose main topic is artificial intelligence, some count any article that mentions AI, some group AI together with categories such as generative AI, deep learning, robotics, or ChatGPT, and some work only in English while others monitor dozens of languages. That is why one database can show hundreds, another thousands, and a global monitoring system far more than either.
Table of Contents
Counting AI news starts with a definition problem
The smallest honest answer is a range. If the count is narrow and curated, the monthly number is already substantial. If the count is broad and global, it rises very quickly. A Frontiers study covering eight newspapers in the US, UK, Germany, and Switzerland found 2594 AI articles from November 2020 to November 2022, which works out to roughly 108 articles a month in that limited outlet set. A German media factsheet found 4968 AI articles from 2018 to 2021, or roughly 103 a month, in its selected sample. A separate study of German-language online outlets found 560 articles on ChatGPT and related text-generating AI between late November 2022 and late April 2023 alone.
Those are not global figures. They are lower bounds drawn from selected publishers and tightly defined samples. Even so, they already show that AI coverage is not a fringe beat that produces a handful of stories now and then. In a relatively small basket of outlets, the category can sustain about a hundred pieces a month before adding wire copy, duplicates, multilingual reporting, blogs, business sites, or the much larger web ecosystem.
Broader trackers change the picture fast
The number jumps once the lens widens. A 2025 preprint built from 20581 AI-related headlines gathered between August 2024 and July 2025 shows monthly counts ranging from 1255 to 2292 in that dataset. That is still not “all AI news on Earth,” but it is a much broader online sample than a handful of legacy outlets.
An even stronger clue comes from the AI News Monitor described by researchers at the Jožef Stefan Institute using Event Registry data. In their March 18, 2024 snapshot, the system showed roughly 1886 articles for generative AI, about 970 for robotics, and about 274 for deep learning on that day’s selected categories. Because those categories can overlap, they cannot simply be added into a clean monthly total. Still, they make one point unmistakable: AI-adjacent coverage can reach into the thousands in a single day once the monitoring is broad enough.
A compact reading of the evidence
| Lens | What the evidence suggests |
|---|---|
| Curated outlet samples | About 100 AI stories a month can appear even in a limited set of newspapers or national outlets. |
| Broader online trackers | About 1255 to 2,292 AI headlines a month appear in one 2024–2025 multi-source dataset, and selected AI subtopics can produce thousands of stories in a single day. |
The first row comes from the Frontiers newspaper sample and the German CAIS media study. The second comes from the 2024–2025 AI headline dataset and the AI News Monitor snapshot.
A worldwide total can only be estimated, not audited
To see why the real global number must be much larger than the narrow studies, look at the size of the news system itself. A 2024 study using GDELT analyzed 140 million news articles from 183 countries in 2022, which is about 11.7 million articles a month. The OECD’s methodological note on Event Registry says the system processes more than 150000 English-language news articles per day, or roughly 4.5 million a month in English alone.
That does not mean AI accounts for a fixed share of all coverage. It does mean any answer in the low hundreds for the whole world is obviously too small. The best reading of the available evidence is this: selected outlet baskets generate hundreds of AI stories a month, broader online datasets generate low thousands, and the true worldwide total for digital articles that mention AI is almost certainly at least in the tens of thousands each month. In busy periods, especially when product launches, regulation fights, chip earnings, or synthetic-media scares collide, the real number could run much higher. That is an inference, not a census, but it is the inference most consistent with the published datasets and monitoring scale.
The moving parts behind the number
The reason estimates drift so widely is not bad math. It is bad comparability. Different systems use different definitions, different source pools, and different units of analysis. GDELT machine-translates coverage from 65 languages and lets users search across that translated corpus. Event Registry classifies articles through keyword and concept logic. The Frontiers newspaper study searched for the full term “artificial intelligence” in print titles. The German ChatGPT study used a narrower query built around text-generating AI. The Slovenian AI News Monitor groups coverage across fields such as generative AI, deep learning, robotics, computer vision, and neural networks.
So the real question is not merely “how many.” It is how many under which rules. Count unique stories and the number drops. Count syndicated rewrites and mirrored business coverage and it rises. Count only English legacy outlets and you get one answer. Count multilingual digital publishing, local press, market sites, trade media, and press-release-fed aggregators and you get another.
The most honest usable answer
How many AI news stories are created each month? The cleanest answer is this: there is no single authoritative monthly total, but the evidence points to a spectrum rather than a mystery. In tight outlet samples, AI coverage already lands around 100 articles a month. In broader online datasets, it lands around 1200 to 2300 a month. In wide global monitoring environments, AI-related categories can spike into the thousands in a day, which makes a worldwide monthly total in the tens of thousands the most credible working estimate, with higher totals entirely plausible depending on how broadly “AI news” is defined.
That is less neat than a single round number, but it is far more useful. A fake-precise figure hides the real story. The real story is that AI has moved from a niche technology beat into a standing global news stream, and the count changes dramatically with every decision about scope, language, duplication, and topic boundaries.
Author:
Jan Bielik
CEO & Founder of Webiano Digital & Marketing Agency
This article is an original analysis supported by the sources cited below
The drivers of global news spreading patterns
Peer-reviewed study using GDELT data that quantifies the scale of global online news in 2022.
https://pmc.ncbi.nlm.nih.gov/articles/PMC10794245/
Event Registry data
OECD methodological note describing how Event Registry classifies AI-related news and stating that it processes more than 150000 English-language news articles per day.
https://oecd.ai/en/eventregistry
Voices and media frames in the public debate on artificial intelligence
Frontiers study analyzing AI coverage in Swiss, German, UK, and US newspapers, including sample size and search design.
https://www.frontiersin.org/journals/communication/articles/10.3389/fcomm.2025.1599854/full
Perception of AI in Slovenia
Conference paper presenting the AI News Monitor and showing category-level counts for AI-related news trends.
https://is.ijs.si/wp-content/uploads/2024/09/IS2024_-_SIKDD_2024_paper_14.pdf
Everything, but hardly any science fiction
CAIS factsheet on German media coverage of AI from 2018 to 2021 based on 4968 news articles.
https://www.cais-research.de/wp-content/uploads/Factsheet-7-Media-Coverage.pdf
Risks and opportunities of generative A.I. How do news media cover ChatGPT
Conference paper studying 560 articles from German-language online news outlets between late 2022 and early 2023.
https://doi.org/10.69931/WFEY9011
The Artificial Intelligence Narrative
Research Square preprint with a 2024–2025 dataset of 20581 AI-related headlines and month-by-month counts.
https://www.researchsquare.com/article/rs-7691931/v1
GDELT DOC 2.0 API Debuts
GDELT documentation explaining cross-language search across machine-translated global news coverage in 65 languages.
https://blog.gdeltproject.org/gdelt-doc-2-0-api-debuts/
