SC Nexus · Weekly
An AI reads the news from around the world each week and picks out the stories that matter most to the most people, using a set of rules anyone can read. It looks at newsrooms in the biggest countries, Google News editions for 50 countries in their own languages, UN, aid and human-rights organisations, and Wikipedia’s day-by-day record of events, and it gives more weight to places where more people live. Luke Cathcart writes the rules and decides what gets published. Every figure links to where it came from. How stories are chosen explains the whole process in plain terms, lists every source, and says what the method gets wrong. What we are following is the list of longer stories the digest keeps tracking from week to week.
Most news is chosen because it sells: dramatic, bad, recent and close to home. That gives a lopsided picture of the world. This digest picks stories a different way. It asks five questions of every story. How many people does it affect? Will it still matter in a year? Is it part of a bigger shift? Is it getting far more or far less attention than it deserves? And how many independent outlets are reporting it? A story is judged the same way whether it is good news or bad.
Every issue ends with its scorecard: every story the AI considered that week, how it answered the five questions, and why. Then a second AI, which has seen only the rules and the scores, checks the marking and says where it disagrees. Those disagreements are printed too, under the original marks. Nothing is quietly corrected. The point is to show the method working, including where it gets things wrong.