Police departments now receive AI-generated crime tips, some completely fabricated. False reports divert limited resources from real cases.
A verification layer that flags inconsistencies in AI submissions by checking against public records, crime databases, and logical plausibility tests. Could integrate with existing tip lines.
Municipal governments or private security firms would pay to reduce false alarms and maintain public trust in AI-assisted policing.
Start with a simple fact-checking API that compares submissions against crime stats and geography data.
Risk: Over-reliance on automated verification missing nuanced falsehoods.