Some police forces score citizens' threat level with Intrado's Beware software based on your medical, criminal history, and social media posts
The new way police are surveilling you: Calculating your threat ‘score’ — FRESNO, Calif. — While officers raced to a recent 911 call …
Context & Ripple Effects
Intrado's Beware gives police dispatchers a threat score for any address or person before officers arrive — built from criminal history, medical records, and social media activity — with Fresno as the visible deployment. The score is computed without the scored citizen's knowledge, making it one of the clearest early examples of consumer data flowing into routine 911 response.
The story reads as a starting point for a pattern the later coverage documents in full: Fusus merging public and private cameras into predictive policing across dozens of cities, ~123 US real-time crime centers combining CCTV, facial recognition, and social media monitoring, and Canadian forces adopting tools like Palantir's Gotham. Beware's core move — scoring individuals from merged datasets — became the template.
First-order effects
- Fresno-area residents are assigned threat scores from medical, criminal, and social media data they never consented to share, and officers respond to 911 calls already primed by that score.
- Officers on the receiving end must act on an opaque vendor algorithm whose inputs and weighting they cannot inspect, shifting judgment from the scene to the software.
Second-order effects
- Vendors follow the demand signal: Fusus scales camera-merging predictive policing to 60+ cities, Benchmark Analytics pitches data-driven identification of problem officers, and real-time crime centers institutionalize the same CCTV-plus-social-media fusion Beware pioneered.
- Data quality becomes the industry's exposed flank — the same challenge that dogs Benchmark Analytics applies to threat scoring, where a stale record or misread post can inflate a citizen's score with no correction path.
Third-order effects
- If the pattern holds, civilian scoring becomes default infrastructure in emergency response, forcing the public-safety AI governance question: who audits the inputs, and what recourse does a scored person have?
- The spread across US cities and into Canada points toward cross-border regulatory attention on predictive policing vendors, with civil-liberties groups positioned as the counterweight to procurement momentum.
The trend: Police technology is converging on automated individual risk scoring built from merged consumer and surveillance data, moving from single-city pilots like Beware to networked crime-center infrastructure.