GCD AI RADAR

GCD AI RADAR

An observatory of AI cybersecurity incidents.

GCD AI RADAR records, counts, categorizes and charts AI cybersecurity incidents so that anyone writing or speaking about the subject has a number they can cite, a definition they can quote and a source they can check. It is an observatory, not a security product: it does not detect, protect or advise.

How it runs

A collector runs every six hours on GitHub Actions, reads feeds, official APIs and public data exports, deduplicates what it finds, classifies it with a language model and commits the result to a public repository. The site is rebuilt from that data. There is no database, no server and no recurring cost; everything is open at github.com/eshin087/gcd-ai-radar. The methodology says what counts and where the numbers are weak.

The gcd family

Licences and reuse

The dataset is CC BY-SA 4.0 (it incorporates AI Incident Database metadata under the same licence) and the code is MIT. Cite the site as “GCD AI RADAR” with a link and the date you accessed it; the data page has downloads and the full attribution block for upstream sources.

Publishers and takedowns

The site stores a headline, a link and a date for each article it cites, plus a two-sentence summary written in its own words. It never stores article text, and it reads feeds and APIs the way they are offered. If you publish one of the sources listed here and would like a title or link removed, open an issue on the repository or contact the maintainer through GitHub; it will be removed.

Privacy

No cookies, no trackers, no accounts. Vercel’s privacy-friendly analytics may count page views; browser Do Not Track and Global Privacy Control are honoured.