Editorial Pipeline Overview

The editorial problem and the data delivery problem are different things.

Editorial problem

There is too much news flow. A pipeline is needed that:

• fetches multiple sources
• removes duplicates
• clusters the same story together
• separates source facts from editorial copy
• shows confidence

Data delivery problem

Once content is built ahead of time, it may not make sense to serve it through a full database server. Then the question is whether mostly-read briefing data can be delivered directly from object storage without a database server.

The role of AI

According to the current report:

• Claude selects the shortlist and final selection
• OpenAI produces the publishable editorial fields
• grounding + publish-gate try to block weak signals
• the public /api/data path does not call AI models on every request