A look at Botometer, which claims to identify the likelihood that a Twitter account is a bot and has been cited by Elon Musk's lawyers in the Twitter lawsuit
In the battle over Twitter's future, the number of bots on the platform is a key issue. Problem is, nobody knows how to count them.
Context & Ripple Effects
The bot question has been the load-bearing wall of the Twitter–Musk fight since Musk made closing contingent on proving bots are under 5% of users, and by the August hearing his lawyers were leaning on Peiter Zatko's whistleblower claims while Twitter conceded its spam counts are only estimates. That admission is what makes Botometer newsworthy: with no authoritative internal number, an academic third-party scoring tool becomes the de facto measuring stick in court.
The deeper problem Wired flags is that nobody agrees on how to count bots at all — a measurement vacuum platforms have left open for years, one that LinkedIn is now trying to close on its own turf with work-email and phone-number profile verification.
First-order effects
- Botometer's methodology — its thresholds, training data, and error rates — suddenly faces adversarial legal scrutiny from both sides, since whichever number it produces could move billions in deal value.
- Twitter's position that spam counts are estimates leaves it arguing in court without a defensible figure, ceding evidentiary ground to any external tool Musk's team can put forward.
Second-order effects
- Platforms outside the lawsuit are reacting to the same trust gap: LinkedIn's new verification and suspicious-message alerts show rivals racing to offer provable human identity rather than estimated bot counts.
- If courts treat third-party scores like Botometer as acceptable evidence, a market opens for auditable bot-detection methods — and for disputes over which method counts, echoing how bot researchers like DFR Lab have themselves been targeted by the accounts they study.
Third-order effects
- If acquisition prices can be contested over a metric the target itself calls an estimate, platform M&A may come to require independent, standardized audits of user authenticity before deals close.
- The pattern points toward a durable industry layer of verifiable-human signals and third-party trust scoring, replacing self-reported platform metrics as the basis for both commerce and litigation.
The trend: Platform metrics are shifting from self-reported estimates to independently verifiable measurements, driven by litigation and advertiser trust pressure.