AI use
Last changed 5 October 2026
Questions? Contact reader@retronewsreader.com
Questions? Contact reader@retronewsreader.com
A note from the developer on where artificial intelligence is at work while you use the reader. There is one such service: Jev, from TypeSafe AI. It is a classifier. It is given a question and a fixed list of possible answers, and it picks one. It writes nothing you read.
What it does
- Sorts stories. Each new story is labelled with a topic and, where one applies, a place, chosen from a fixed list. The Topic and Place filters and a story's tags run on these labels.
- Works out what you meant. When you add an outlet by typing its name, it picks the most likely match from a short list, so a misspelling still finds the outlet.
- Chooses among an outlet's feeds. Large outlets publish dozens. It picks out the main ones to offer you.
- Finds the support link. It picks which link on an outlet's home page is the one for donating or subscribing.
What it is sent
- To sort a story: the outlet's name, the headline, the summary, the first 600 characters of the story and the publisher's own categories. All of it is already public.
- When you add an outlet: the words you typed into the lookup box, the outlet's web address, and the names and addresses of its feeds and home page links.
- Never: your screen name, email, password, what you read or saved, your friends, or your messages. No request says who is asking.
What it does not do
- It does not write, summarise or rewrite anything. Headlines and story text are the publishers' own words, shown as published.
- It does not choose or rank what you see. The list is in order of time, from the outlets you follow. Nothing is recommended and nothing is personalised.
- It does not read the Gossip Column. Messages are sealed in your browser before they are sent, and neither the server nor any AI service can open them.
- It is not a chatbot, and this project does not use anything you do in the reader to train one.
Where it can be wrong
Topic and place labels are a best guess and are sometimes wrong. A mislabelled story is still in your list: it is only filed under the wrong tag. If the service cannot be reached, stories arrive without labels and everything else works as usual.
What TypeSafe AI does with what it is sent is in its own terms and privacy policy.
What the reader keeps about you is on the privacy page.