The Top 7 shows what a balanced panel of 30 big newsrooms is covering most, ranked mechanically by coverage. An AI then gives each story an Impact Score, and picks Quietly Important, stories with impact but little buzz, by rules Seth set. Here are the rules, in plain words.

1. The Top 7 is ranked by coverage, computed mechanically. Each edition a script scans the homepages and main feeds of a fixed panel of 30 English-language news outlets (listed below) and adds up how prominently each story appears. No person or AI picks the order: the 7 stories with the highest coverage scores, weighted as below, are the Top 7.
2. Why 30 outlets, built this way. The panel starts with the 15 most-read news outlets, kept whatever their lean, because they are what most people actually see. Those 15 lean left on balance, so we add 15 more: the next most-read outlets on each side, mostly right-leaning, until the panel has as many Left and Lean Left outlets as Right and Lean Right ones. Audience weighting still applies, so the biggest outlets still count most; the tally below shows how that balances out.
3. Bigger audiences count for more. Each outlet is weighted by its monthly audience, using published Similarweb figures reported by Press Gazette. Where no published figure exists we use a rough, labeled tier instead of inventing a number.
4. Placement counts too. A story an outlet leads with scores 3, one in its top block scores 2, and one further down scores 1. A story's coverage score is the sum, across outlets, of audience times placement.
5. Social media counts, with a cap. We count recent posts about each story on X and convert them to estimated readers at a fixed, deliberately rough rate (about 1,000 views per post). That estimate is capped so social buzz can't outweigh the newsrooms. Reddit's main news boards are counted when we can reach them.
6. Separately, we rank impact. Each edition our AI editor also picks the 10 fresh stories of the day most likely to have a lasting, long-term effect on the world, ranks them #1 to #10, and gives each an Impact Score out of 10. This is the AI's judgment, made by rules Seth set, not coverage: how many people it touches, how badly, for how long, and whether it can be undone.
7. Impact Scores are calibrated hard. 10 is a story that happens a handful of times a year, if ever: a major war starting, a pandemic declared, a market crash, a regime falling. 8-9 is one of the big structural events of the year. 5-7 is a meaningful development with lasting effects. 3-4 is real but modest or reversible. 1-2 has little lasting effect. Rank and score are independent: on a slow day #1 might score 4.5; on a big day several stories might score 8 or more.
8. Every Top 7 story is tagged. If it is on the impact list it reads like "#3 in AI Impact (7.4/10)". If not, it is marked "Buzzing": widely covered, but not among the day's most consequential, with its AI Impact Score shown for honesty. Every score you see is the AI's, not a human editor's.
9. Quietly Important is the rest of the impact list. Every impact-list story that didn't make the Top 7 appears there in impact order, as picked and scored by our AI editor: developments with consequence but not a lot of buzz.
10. At least five Kingdom stories, scored honestly. Kingdom stories cover the global church and religious freedom: persecution, martyrdom, imprisonment, religious-freedom laws, denominational decisions, church growth, and major changes for hugely influential Christian figures. Each edition carries at least five, wherever they land. If fewer make the impact list, extra ones are added at the end of Quietly Important with an Impact Score but no rank. They are scored by the same rubric as everything else, even when that puts them at the bottom. Each is marked KINGDOM next to its region.
11. Our one stated bias. The publisher, Seth, is a Christian and has told our AI editor to follow the persecution of the global church closely. The AI scores those stories by the same rubric as everything else, so the bias shows only in their presence on the page, not in their scores.
12. Two sources, balanced. Each Top 7 story links two sources: either two rated Center, or one leaning left and one leaning right, using the ratings shown below. Each link shows its lean.
13. Fresh, sourced and neutral. Stories must have broken or moved in about the last 24 hours; older items carry their dates. A story from an earlier edition returns only when something real has changed, marked "Update:". We don't take sides, sensationalize or invent facts, and market figures come only from sources checked that day.

Tally: 12 Left or Lean Left, 6 Center, 12 Right or Lean Right. By audience, roughly 46% left, 36% center, 18% right, because the most-read outlets carry the most weight.

Lean ratings: AllSides Media Bias Ratings (online news). Audience figures: Similarweb via Press Gazette, or labeled tiers. Ratings for the 10 outlets added in November 2026 come from AllSides' published chart and will be re-checked at the next review. Last reviewed November 2026; sources and ratings are reviewed every six months.

Each edition is put together twice a day, around 7 a.m. and 4 p.m. Central, by an AI news editor working from these rules, under Seth's direction.