About Code Word
Our parent site, AGI Scorecard, tracks how far AI capability has actually come. This site handles what follows from that: once a voice and a face can be generated on demand, how do you verify anything. It is a judgement desk, not a scare-story blog.
Method
| Every judgement page carries | The question as the title · a one-line verdict above the fold · a table · visible FAQs identical to the structured data · a dated primary source with a link · the conditions that would flip the verdict |
|---|---|
| Tools | Run entirely in your browser. Checklist and scam-check results are written into the URL so they can be shared; the code word itself is never uploaded and never written into a link — that page has no network call at all |
| What we leave out | Loss statistics we cannot trace to the publishing agency's own report · scare-story case studies · named "scam apps" and blocklists · any paid product, course or recovery service |
Verification, stated plainly
Some government and research domains are unreachable from the environment this site is maintained in. Where that applies, we link the primary source so you can check it directly, and we only state figures and dates that several independent, named professional sources report identically. If a number appears here without a traceable source, that is a bug — the rest of the site is built so it cannot happen quietly.
Limits
- Not legal advice. Rules are summarised; the linked originals and the enforcing authorities govern.
- Not a substitute for reporting. If money has moved, your bank and the official reporting route come first.
- US and EU/UK focus. Reporting routes outside those are not covered rather than guessed at.
- No account, no IP address stored, no full browser fingerprint.
Network
agiscorecard.com (AI progress, scored) · EU AI Act ledger · 识骗 (the same ground under Chinese law) · The Power Bill (is AI raising your electricity bill?).
Machine-readable notes: /llms.txt · full text: /llms-full.txt · source: github.com/f-tiger/agi-site