How Bynamely uses AI

A model is called in exactly two places in the whole pipeline.

Sorting the ambiguous

Rules settle many headlines on their own. The ones they cannot call — a name in a listicle, a passing mention — go to a model, which answers whether it is real news about that person, and what kind: an interview, a casting, a release.

Writing the summary

One call per update, never per article. The model is shown the attached reports and nothing else, and writes the headline, a short summary, a line on why it matters and a few key facts.

These are the decisions that determine what reaches you, and no model makes them. They are rules — the kind you could work through by hand and get the same answer.

• Who a report is about. Exact name matching, then fuzzy string comparison against the registry. No model is asked to identify anybody.
• Which reports are the same story. A weighted score over shared names, headline similarity and timing. Deliberately not semantic similarity.
• How important something is. A short fixed formula: a baseline for the kind of event, plus a step for each additional outlet reporting it, plus a bonus if one of them is the person’s own channel. The kind is the one input the model touches — rules read it from the headline, and the model’s sorting answer is used first — so it can move a story from “interview” to “casting”. The number itself is never the model’s.
• Whether it is confirmed, reported or unverified. Four rules over the evidence — how many outlets, how reliable, whether any is official. It never sees the story, only the shape of the sourcing.

That split is the point. If a model scored importance, an important-sounding rumour would get help from sounding important. Keeping the two apart is why a rumour can be marked urgent and still never be sent.

Before any summary is published, every name, number and date in it is read back against the reports the model was shown. Anything not found there is, by definition, invented.

It is all-or-nothing, and it never edits. One unmatched name and the entire summary is discarded — not softened, not rewritten. The update keeps whatever headline it already had and the reports stay attached, because they were always the record.

The check is string comparison, not a second model. Asking an AI whether another AI made something up is expensive and circular: it is the same kind of system, with the same failure, being asked to audit itself. Names, numbers and dates are where fabrication actually shows up, and those can be checked mechanically.

A summary is not a quotation, so the check has to tolerate the same fact written differently. It treats ’90s and 1990s as one claim, accepts 56 where an outlet wrote 56th, and lets a name be spelled out or abbreviated in either direction.

What it refuses is a claim getting sharper. If the reporting says ’90s, nothing licenses the model to write 1994 — naming a year is a stronger statement than naming a decade, and the evidence did not make it.

It also refuses near-misses on names, and the clearest example is one where the model was right. An outlet misspelled an actor as Prithiviraj; the summary corrected it to Prithviraj, and the check threw the summary away. That is annoying and it stays, because close-enough name matching is the exact hole a real fabrication walks through — two different people are often one letter apart.

Measured on everything published so far, not projected, and read live from the same endpoint the front page uses.

6180 of 6330

published updates carry an AI summary that passed the check

0

rumours have ever been sent as an alert

150

published with no summary, because nothing we wrote stayed inside the evidence

The last of those is the one worth reading twice. When a summary cannot be written inside the evidence, the update ships without one and says so — you will see “nothing we could write stayed inside what the reports actually say” in place of the prose. An empty summary is a check that worked.

The model is shown the attached reports and the names of the people the update is about. That is the entire prompt.

It is never shown who you are, who you follow, what you have read, or anything else about your account — there is no path in the code from a reader to a prompt. Nothing you do here trains any model, and we do not build one.

Because the job is compression, and there is too much of it to do by hand. One story about one person is often filed by several outlets within the hour, each with its own headline. We follow over a thousand people across film and sport, around the clock. Turning each of those stories into one plain, sourced update is what a model is good at — and it is the only way a small team can do it for everyone, not just for the most famous few.

The alternative was to show you the outlets’ headlines as they come. That tells you less, repeats itself every time a story is re-filed, and copies other people’s writing. A model’s summary under a mechanical check tells you what happened once, in plain words, with every report one tap away.

Updates are published as they arrive, without a person reading them first. What stands between the model and you is the check above, which runs on every one, and the confidence rules: an update resting on one weak source is labelled unverified, kept out of search engines and never emailed.

People come in afterwards. A correction or a withdrawal is made by a person, with a written reason, and the update itself says it was corrected or withdrawn.

We use more than one AI provider, so that when one is unavailable the same work goes to another rather than an update going unwritten. Which one answered is recorded against every summary.

The check above applies identically whichever answered. A change of provider cannot loosen what a summary is allowed to say.

The check catches invention, not every possible error — a summary can be faithful to the reporting and the reporting can still be wrong. When that happens we correct or withdraw the update at its original address rather than deleting the page, so a link we emailed you takes you to the withdrawal instead of a 404.

Every update lists the outlets it was built from, by name and by link, so you can go and read them. More on how we source and correct.