Facial Recognition and Law Enforcement

Introduction

• NYPD used facial recognition to catch a shoplifter based on a vague description resembling Woody Harrelson.
• Raises questions about the appropriateness of police use of facial recognition technology.

Georgetown University Study

• Study revealed misuse of facial recognition systems by police departments.
• Some departments altered facial features in images to generate matches (e.g., changing eye shapes, mouth positions).
• Lack of strict rules governing the use of algorithms leads to arbitrary police stops.

How Facial Recognition Works

• Core concept: tracking key facial landmarks (e.g., distance between pupils, nose angle, cheekbone shape).
• Best results from straight-on photos (e.g., passport, driver's license) with at least 80 pixels between pupils.
• Sophisticated programs can recognize features at angles or with partial obstructions as long as crucial features are visible.

Vendors and Availability

• Major vendors: NEC, Morpho, Cognitech.
• Amazon and Google have integrated facial recognition into their cloud services, making it widely accessible.
• Almost anyone can create a facial recognition system with basic coding skills.

Accuracy and Standards

• Programs operate on accuracy thresholds; however, no firm rule dictates the required match score.
• Police can adjust thresholds to classify a match, leading to potential wrongful stops based on vague resemblances.

Concerns with Racial Bias

• Algorithms show higher error rates for women and people of color, often due to biased training data.
• Government testing consistently reveals higher false match rates for Black individuals compared to white individuals.
• This raises concerns about targeting marginalized communities.

NYPD and Arrest Statistics

• NYPD claims no arrests solely based on facial recognition; however, technology has been involved in over 2,800 arrests since its implementation.
• False matches can lead to police stops, which pose dangers even without arrests.

Community Reactions and Controversies

• Advocates argue facial recognition helps police protect communities (e.g., Detroit's Project Greenlight led to a 23% crime reduction).
• Critics point to a lack of transparent oversight and overwhelming police response to new tips.
• Controversies led to heated discussions, including the arrest of a police commissioner opposing the technology.
• Some cities, like San Francisco, have banned police use of facial recognition, citing privacy and oversight concerns.

Conclusion

• The issue extends beyond technology; it raises profound questions about government trust and oversight in the use of surveillance tools.
• Need for deeper discussion on the implications of powerful surveillance technologies in everyday life.