How Boring works

The sources

Boring reads the public RSS feeds of 70 outlets: 22 general news outlets, which make the main edition, and 48 specialist outlets for the app’s subcategories. From each article it keeps only the headline, a snippet of at most 400 characters and the link, never the full text, and only articles from the last 36 hours.

Political leaning

An outlet’s political leaning, left, center or right, comes from independent media-bias ratings, never from Boring’s AI. An outlet no rater covers is marked “Not rated” and counts for neither side. Pages show the ratings as letters (L, C, R), never as colours.

Until Boring licenses a rater’s data, the ratings are provisional placeholders set to match public ratings from AllSides and Ad Fontes Media, and every story page says so.

Choosing the stories

An event makes the edition only if at least three general outlets covered it, including both a left- and a right-leaning outlet, or a center outlet. Events are ranked by how widely they were covered. Never by clicks, shares or time spent. The edition currently takes the top 10, and then it ends. A story from the previous edition comes back only when there is something new to say.

The role of AI

Claude, an AI model made by Anthropic, writes each story from what the outlets reported, and checks it against those reports before it is published. Every fact links to the articles that report it; anything that can’t be matched to them is left out. It doesn’t rank the stories or decide an outlet’s political leaning. Because Claude is made by Anthropic, a story that mentions Anthropic says so.

Plain words

No loaded words such as “slams”, “shocking” or “historic”, no exclamation marks, no speculation, and nothing longer than it needs to be.

How it was covered

Each story has one line on how left-leaning outlets covered it and one for right-leaning outlets: what each emphasised, included or left out. When outlets across the spectrum reported it the same way, the story says so. When there isn’t enough coverage from both sides to compare, it says that instead.

It describes coverage, never who is right, and it never invents a difference.

Primary sources

Where they exist, a story can link the documents behind it, such as a bill, a court filing or an agency release, from official sites only. This is currently off while Boring is new, so most stories have none.

What changed

When a story returns from the previous edition, it lists what’s new since it last ran, each with its sources, and the edition marks it “Updated”.

Background

Background explains the cases, laws, programs and earlier events a story names, from official records and earlier reporting, with every source linked. It is labelled as background, not as the day’s coverage. When it is revised, the story lists what changed under “How the background changed”.

The show

Every Saturday the week’s most widely covered stories become a short show, free on the Listen page and as a podcast. Two AI voices read it, made only from the stories already published.

Reader polls

Each Saturday’s email asks a few questions about the week’s top stories, about readers’ own views, never about facts. Results on the Polls page give percentages, and breakdowns by age, region, political leaning, gender and marital status, one at a time and only for groups of at least 30 answers. They are always labelled as Boring readers who answered, not a representative sample.

The weekly vote

Subscribers also vote each week on what Boring builds next, from two to four small candidates. Totals stay hidden until voting closes, so early results can’t sway it; then they are published, and the winner ships within the week.

Corrections

Published editions are never rewritten. When a story gets something wrong, a correction records the text as published, the corrected text, a short note and the sources it rests on, and the Boring team approves it before it is published. The story’s page then shows the corrected text with a dated note, every correction is listed on the Corrections page, and the next weekly email and show announce it.

To report a mistake, use the feedback form: every email from Boring links to it, and it can email the link to any subscriber. Or write to boringnews.hello@gmail.com.

What Boring never does

• Store or show an article’s full text. Only headlines, snippets and links.
• Publish a fact without a source.
• Rank stories by clicks, shares or time spent, or run an endless feed.
• Decide an outlet’s political leaning itself.
• Say which side is right, or colour the sides red and blue.
• Rewrite a published edition.
• Show ads or track readers.

Limitations

• AI makes mistakes. The checks catch many but not all, and when the sources get something wrong, Boring can too.
• Boring sees only what outlets put in their public RSS feeds, from the last 36 hours. A feed that fails one morning is missed that day, and outlets without a feed are never read.
• Each story is written from a balanced sample of up to 12 articles, not every article.
• The political-leaning ratings are provisional until Boring licenses a rater’s data, and many specialist outlets aren’t rated at all.
• Wire copy counts more than once: an AP or Reuters story republished by several outlets counts as several outlets’ coverage, which can make a story look more widely covered than it was.

Provenance

Every story records how it was made, shown on its page under “How this story was made”, and every published edition is saved as it was and never overwritten.