Paste a post from X or YouTube. OmiSphere pulls the comment section, you pick the accounts to check, and each one comes back with an OMI score and a plain-English read: a real person, or a bought account. Every verdict shows the behavioral evidence behind it.
How it works
Drop in any X post or YouTube video. OmiSphere pulls its comment section — up to every account on the post.
Read through the comments and select the ones you want checked. You spend credits only on the accounts you choose.
Each account comes back with a 0-to-100 OMI score and a plain-English reason: real person, or bought, and the evidence behind the call.
What you get
Every account you check gets its own 0-to-100 score for how likely it is bought or inauthentic. No blanket verdict on the whole comment section — each account is judged on its own evidence.
The omi analyst reads the evidence and tells you, in a sentence or two, why an account looks real or bought. No jargon, no black box.
Posting cadence, repetition, writing tells, profile metadata, personal-voice rate, engagement farming, account history, and fingerprint memory. Computed, not guessed — fast and auditable.
OmiSphere pulls the whole comment section for free. You choose the accounts worth checking, and credits go only to the ones you select.
Every scan adds a behavioral fingerprint to the database. Later scans pull priors from what came before, so repeat offenders surface faster.
Open any score to see the raw signals that drove it — the comment, the account metadata, the cadence. Every number shows its work.
Scope, plainly
Today OmiSphere reads X (Twitter) and YouTube — posts, videos, and the accounts that comment on them. The detection engine is platform-agnostic; Reddit and TikTok are next. We would rather ship two platforms with depth than four with stubs.
Start your first scan
$9.99/month · 20 credits monthly · 25 free to start. Cancel anytime. Built for creators, brands, journalists, and platform-integrity teams.