Our standards

Trust is a practice.

These are the rules we use to earn it — story by story.

01

Evidence before excitement

We separate company claims from independently established facts, link to primary material and explain uncertainty.

02

Plain language

We make complex ideas approachable without pretending they are simpler than they are.

03

People remain accountable

Automation can help collect, compare and check information. A human editor remains responsible for every published story.

04

Visible correction

When we make a meaningful error, we correct it promptly and add a clear note describing the change.

Sourcing and fairness

We prefer original documents, official datasets, research papers and direct, on-record interviews. We identify advocacy reports as advocacy reports and company claims as company claims. When a person or organisation faces serious criticism, we seek a meaningful opportunity for response.

How we use AI

AI tools may support discovery, transcription, translation, comparison and preliminary drafting. They do not publish autonomously. Every article is checked against its cited sources by a human editor. We do not invent quotes, sources or reporting. Generated artwork is labelled as an editorial illustration.

Corrections

Factual errors are corrected in the article. Material corrections receive a dated editor’s note explaining what changed. Typos or style changes that do not alter meaning may be fixed silently. Send correction requests to contact@modelcurrent.com with the article URL and supporting evidence.

Commercial independence

Advertising and sponsorship do not buy favourable coverage or advance access to editorial conclusions. Sponsored material is clearly labelled. Affiliate links, if introduced, will be disclosed near the link.