One story. Every side. No mastheads.

The pipeline

Read. Briev continuously pulls reporting from 98 newsrooms spread deliberately across the political spectrum — left, centre and right — plus wires, trade press and science desks.

Cluster. Every article is embedded and matched against the others, so the forty separate write-ups of one event collapse into a single story with forty sources behind it.

Write. Briev then writes its own account of that story from the whole cluster. Nothing on this site is a copied paragraph: the summaries, the headline, the "why it matters" — all of it is generated from the underlying reporting.

Compare. For stories carried on both sides, Briev compares how each camp frames the event, what each pushes to the front, and what only one side mentions. That comparison is the product.

What we flag

Blindspots. A story one half of the spectrum is running hard while the other half stays quiet. The flag is about who is covering it, never about whether the story is true.

Underreported. Stories Briev rates as significant that almost nobody is carrying.

Cross-spectrum differences. Same event, materially different accounts.

Coverage tilt. Whether one side is publishing above its own normal pace — measured per outlet against its own baseline, because raw volume always flatters whoever publishes most.

Why there are no publisher names

Briev reports on coverage, not on newsrooms. Naming mastheads turns every finding into a fight about the outlet instead of the pattern, and a single article is never a fair verdict on a publication. So the site speaks in camps — left, centre, right — and in counts.

For the same reason there are no bylines and no links out to the original articles. What you read here is Briev's own account, and the political-lean ratings behind the aggregates are coarse, public, third-party ratings applied at the outlet level, not per article.

Limits

This is automated analysis and it can be wrong. Lean ratings are approximations. A blindspot can simply mean a story broke an hour ago. Framing comparisons are generated from the text of the coverage, so they inherit its blind spots as well as describe them. Treat everything here as a map of the coverage, not a verdict on the world.