89% single-voice: how layoff news is really made

832 of 937 layoff stories, 89%, had a single voice in the week from 7 to 13 November 2026, according to LSN News, my news engine, which searched eight languages. And the stories that framed job cuts as an “AI” story did something stranger: they reached 25 to 52 outlets from only 2 to 5 independent voices. Many outlets, few sources. Here is how that happens, why it is usually innocent, and a simple test you can use the next time a layoff headline seems to be everywhere.

Contents · 10 sections

TL;DR #

• In one week (7–13 November 2026), 832 of 937 layoff stories in LSN News had a single voice.
• Almost all of those are local stories one outlet wrote and nobody else picked up. That is normal.
• Three AI-framed job-cut stories reached 25 to 52 outlets from just 2 to 5 voices, through one network each (ACM, Big News Network, MediaNews).
• Many outlets is not many sources. Count voices and primary sources, not websites.
• The data covers about one week, so it shows a pattern, not a trend.

Two kinds of “single voice” #

When I first saw 89%, I assumed syndication: one article copied onto many sites. That is not what most of it is.

A re-count on Friday evening, with a stricter word list, gave the same picture (495 of 545 layoff stories with one voice, about 91%). But almost all of those, 488, were also single-outlet stories. One newspaper wrote it, nobody else picked it up.

That is the normal shape of layoff news. It is local: one employer, one town, one reporter who knows the plant. Some examples from the week:

• “Saratoga Council urged to reconsider layoffs”
• “499 jobs lost: Retired nurses protest Hamilton hospital cuts”
• “Homag baut Stellen ab” (German machine-maker cuts jobs)
• “Carl Zeiss Meditec Stellenabbau Oberkochen”

None of these travel, and they shouldn’t have to. But it means that the picture of “the layoff wave” you get from national and international news is built from a few dozen stories out of hundreds. The rest is a quiet background of local cuts.

The few that travel on their own #

At the other end are the layoff stories that many newsrooms wrote independently. In LSN News the biggest of the week were:

• Qantas cutting jobs and sending roles overseas: 156 outlets, 132 independent voices, 18 languages (LSN News).
• Boeing cutting 5% of its commercial plane workforce: 178 outlets, 99 voices.
• Vodafone cutting 3,000 jobs in Europe: 223 outlets, 78 voices.
• PepsiCo, 1,000 North American jobs: 45 outlets, 40 voices.

What they share: a large named employer, a hard number, and an official announcement. Each newsroom can check it and write its own copy. That is what independent coverage looks like.

The ones that travel on a network #

Now the interesting cases. Three stories that framed job cuts around AI:

• “Bank bosses face inquiry over AI and offshore job cuts”: 52 outlets, 2 independent voices. Every one of the 52 outlets belongs to one network of Australian regional newspapers, Australian Community Media (LSN News).
• “US-China AI race could face financial setbacks, mass layoffs over next three years: Report”: 25 outlets, 2 voices. 23 of the 25 are sites of one aggregator, Big News Network.
• “What to consider if company making AI-related job cuts wants you back”: 45 outlets, 5 voices. 34 of them belong to MediaNews Group’s family of US papers.

Count outlets and each looks like a big story. Count voices and each is one or two pieces of writing, republished by a group that owns or feeds many sites. In all three, one or two headlines, word for word, cover nearly every copy.

Compare that with the German labour agency chief’s warning that AI will eliminate 800,000 jobs: 82 outlets, 58 independent voices. That one is genuinely widely reported. So the “AI and jobs” theme is not only syndicated. But the specific stories that made it look like a flood this week mostly were.

What syndication is, in plain words #

Syndication is how most local news works today. A newspaper group owns dozens of regional titles. It writes a story once, at a central desk, and publishes it on every title, sometimes with a local headline. An aggregator does something similar with other people’s copy: it buys or licenses stories and places them on many branded sites. News agencies sell the same article to thousands of customers.

None of that is sinister. It is cheap and it keeps small papers alive. The Australian group that carried the bank story also carried 67 copies of the Microsoft outage coverage the same week.

But it changes what a count means. 52 outlets sounds like 52 newsrooms decided a story mattered. In reality one newsroom decided, and 51 websites displayed the decision.

In my library, Walter Lippmann noticed this a hundred years ago, in Public Opinion. Describing how editors under deadline borrow the shape of each other’s coverage, he wrote: “It is here that newspapers influence each other most deeply.” Back then it was cable copy from London. Today it is a content management system pushing one article to fifty domains in a second.

Why “many outlets” doesn’t mean “many sources” #

There are three different numbers hidden behind “everybody’s reporting it”:

1. Outlets: how many websites carried it. This is what search results and social feeds show you.
2. Independent voices: how many different pieces of writing there are. LSN News groups outlets that carried the same headline, and treats each known network as one voice.
3. Primary sources: who actually knows. The company, the regulator, the union, the court filing.

A story can be big on the first number and tiny on the other two. That is exactly the pattern of the AI-layoff pieces above. The bank story did lean on real sources, the banks’ own results and the Financial Services Union’s estimate of more than 10,000 banking jobs gone in two years, but only two newsrooms wrote about it. The US-China story was one Indian news-agency piece summarising one S&P Global Ratings report, and one aggregator made it look like 25 papers agreed.

Checklist: a test you can use #

Next time a layoff story, or any story, seems to be everywhere, try this. It takes about two minutes.

1. Open three of the articles from different-looking sites. Are the first two paragraphs identical? Then you are reading one article several times.
2. Look at the small print. “This article originally appeared in…”, “Distributed by…”, a group logo in the footer, or the same byline on every site. Those are syndication marks.
3. Find the primary source. Who announced the cuts? A filing, a press release, a union, a minister? If no article names one, the number in the headline is someone’s estimate.
4. Look for a second newsroom. Is there one article, from a different owner, written in different words, that checked the same facts? If yes, the story has at least two voices. If not, it has one, however many sites carry it.
5. Check the date. Syndicated stories often keep circulating for days after the news moved on.

On LSN News, the story page does steps 1 and 4 for you: it shows how many outlets carried a story and how many independent voices those outlets represent.

What this data can’t tell you #

I want to be careful here, because “few voices, many outlets” is the kind of finding that gets turned into something it isn’t.

• About one week of data. LSN News started collecting on 7 November. Everything here is one week, not a trend. I can’t tell you whether layoff news is more syndicated than it was last year.
• Word patterns miss things. I searched eight languages for layoff words. Some stories use other words, and I removed sports “playoffs” by hand.
• “Independent” is measured, not proven. Two outlets that rewrite the same agency copy with different headlines count as two voices. So the true number of independent voices is, if anything, lower than shown.
• Syndication is not intent. These numbers show that one network republished one piece. They do not show why. Republishing deals, editorial calendars and simple cost-cutting explain them fine. I come back to that question in a separate post.

Frequently asked questions #

What does “single voice” mean? #

It means all the outlets that carried a story ran the same piece of writing, or only one outlet ran it. LSN News groups outlets with the same headline, and outlets in one known network, into one voice.

What is news syndication? #

A newspaper group, agency or aggregator writes or licenses a story once and publishes it on many websites. It is a normal, cheap way to fill local papers, not a sign of manipulation on its own.

Are AI layoff stories being pushed by the media? #

This week’s data shows three AI-framed stories that reached many outlets through one network each, with 2 to 5 independent voices. It shows republication, not intent, and one of the three argues the opposite of the “AI takes jobs” frame.

How can I check if a story is really widely reported? #

Open three copies from different-looking sites, look for syndication marks, find the primary source, and look for one article from a different owner in different words.

The lesson #

Layoff news is mostly made in two ways. Most of it is local and alone: one paper, one employer, one town, read by the people who live there. A small part is amplified: one desk writes it and a network multiplies it.

Neither is a problem in itself. The problem is when you count the second kind as if it were consensus. When an “AI is taking the jobs” story seems to be everywhere, it is worth asking how many people actually wrote it. This week the honest answer was often two.