Free check
One assistant, three questions, one sample each, a short score, and a result page you can share.
Free, with a verified email.
What AI assistants say about you
Find out what ChatGPT, Gemini, Perplexity, and Claude say about you.
Saidbrief asks four assistants about you and quotes what each one said. It scores the answers and writes a plan to correct the sources. The free check takes two minutes.
Three steps, and you see the words.
-
You tell us who you are
Your name, your profession, your city, and a few facts that are true about you. That profile is the ground truth every answer is checked against.
-
We ask four assistants, several times each
Ten questions, from "Who is {name}, the {profession} in {city}?" to "Who are the best {profession}s in {city}?" Each assistant answers through its developer interface with web search on, three times, because answers vary.
-
You read what they said and what to do
Every answer is quoted with the date and the model. Five scores say how you came out. A plan says what to correct, where, and why it helps.
Five scores out of 100, each with its reason.
-
Recognition
Do the assistants know who you are when asked by name?
-
Accuracy
Do the facts they give match the facts you gave us?
-
Identity
Do they keep you apart from people with your name?
-
Visibility
Do you appear when someone asks for the best people in your field and city?
-
Sources
Do they cite your own website and profiles, or someone else's?
Each score shows its spread. If ChatGPT recognized you in two of three tries, the reading shows it.
Anyone who sells their own judgment under their own name.
Clients now ask an assistant about a person before the first call. The assistant may describe you well, describe a namesake, give your old firm, or say nothing. The reading shows you which.
It works the same way for physicians, speakers, authors, and architects.
One reading, or a watch on your name.
Full Reading
Four assistants, ten questions, three samples each, and all five scores. Every answer quoted, the sources, the correction plan, a PDF, and a re-check 30 days later.
$29, once.
Monitor
A re-check every month on the five core questions, a summary of what changed by email, and your history.
$12 a month.
Agency
Up to 25 readings a month, a white-label PDF with your logo and contact, a client list, and Monitor for every client.
$79 a month.
A dated record, shown as it was.
Each reading shows what each assistant said when we asked on that date, through its developer interface with web search on. The consumer apps can answer differently from their interfaces, and any assistant can answer differently from one day to the next. That is why we ask three times and show the spread.
No one controls what an assistant says, and Saidbrief does not claim to. The correction plan improves the odds by fixing the sources the assistants read.
Three questions, answered.
Is my result public?
No. Results are private until you choose to share one. A shared page shows your name, profession, city, scores, and the quotes, and never your email.
Do you keep my profile?
Yes, while your account exists, so a re-check can compare. You can delete any check or your whole account from the account page. We train nothing on your data.
Why three samples?
Because the same question gets different answers from the same assistant on the same day. Three samples show whether a result is steady or a coin toss.
Built by Brett Wharton, Founder of NexusMKTG in Boston, who has built 200+ websites and apps. He made the first reading on his own name.