Flock CEO Garrett Langley on Controversy, "Surveillance State" Claims, and Privacy vs Safety

Episode Summary

Executive Summary: The conversation centered on Flock’s role in public safety vs. privacy, with Garrett Langley defending license plate readers and drones as community-controlled tools that solve crimes and find missing people. He emphasized stricter defaults, shorter retention, stronger audits, and transparency portals while acknowledging mistakes, PR fallout, and the need to better communicate limits—especially around facial recognition, video capture, and AI.

Main Topics: Safety vs. privacy tradeoff (Priority: 5/5): The core debate was how much privacy communities should surrender in exchange for faster crime solving, missing-person recovery, and deterrence. Both sides agreed the issue is nuanced and should be decided democratically at the local level. How Flock works and what it records (Priority: 5/5): Langley explained Flock’s license plate readers, what data is captured, what is not captured, and how retention periods affect effectiveness. He stressed that Flock does not do facial recognition, capture video, or look inside cars. Community governance and retention policy (Priority: 5/5): The discussion focused on how city councils approve deployments, set retention windows, and decide who can access shared data. Langley said Flock moved its default retention from 30 days to 7 days while preserving local choice. Abuse prevention, audit logs, and bad cops (Priority: 5/5): A major segment addressed the risk of internal misuse by police. Langley described a new Audit Assistance tool designed to flag abnormal searches and help identify officers abusing the system. Public backlash, misinformation, and PR recovery (Priority: 4/5): The interview explored vandalism, social-media-driven opposition, and cases where cities removed cameras because of mistaken beliefs about features such as facial recognition. Langley said the company must improve communication and accept responsibility. Drones, transparency, and AI (Priority: 4/5): The conversation expanded to Flock’s drone business and the risks/opportunities of AI in public safety. Langley favored human-in-the-loop systems, delayed transparency, and cautious adoption rather than broad predictive surveillance.

Key Arguments: Local communities, not the vendor, should decide the acceptable balance between safety and privacy through elected officials and city councils. Data retention is a direct tradeoff: shorter retention reduces utility, but most crimes can still be solved within a week of data. Flock’s product is narrower than critics assume: it does not use facial recognition, does not record full video, and does not search for people on licensed cameras. Abuse by law enforcement is real and should be proactively detected with stronger audits, not left to manual spot checks. Companies selling public-safety tools must own their downstream impacts, set strong defaults, and communicate clearly to the public. AI should be used cautiously in policing, with humans in the loop and third-party validation, because errors can have severe consequences. Drones can improve safety by resolving ambiguous threats faster and reducing dangerous officer responses, but they also require transparency and safeguards.

Data Points: Years since Flock started: 9 years - Langley said he started the company nine years ago with a mission to live in a safer neighborhood. Communities using Flock: roughly 6,000+ cities - He said Flock is deployed in just over 6,000 cities across the country. Crimes solved: north of 1,000,000 - Langley said more than a million crimes were solved using Flock last year. Missing people found: north of 10,000 - He said Flock helped find over 10,000 missing people. July missing-person cases: about 1,125 - He cited July as having around 1,125 missing people recovered. July stolen cars: over 20,000 - He cited more than 20,000 stolen cars in the same period. Default data retention previously: 30 days - Langley said 30 days had long been the default retention period. Current default data retention: 7 days - He said Flock recently shifted its default retention to seven days. Estimated crimes solved within 7 days: 90% - Langley estimated that 90% of crimes would be solved within seven days of data retention. States with five-year retention: New Jersey - He said New Jersey state law requires five-year retention. State max retention example: California and Washington have max limits - He noted several states now cap retention by law. Approximate local government meetings attended: close to 10,000 city council meetings last year - He described Flock’s sales process as heavily dependent on city council approval. Audit Assistance rollout: built 4 months ago - He said the new internal abuse-detection tool was built four months prior. Detected abuses over 8 years: just under a dozen - Langley said Flock had detected just under a dozen abuses over eight years before the new tool. Large-agency example: about 1,000 officers a day - He used a Southern California agency to illustrate why manual audits are insufficient. Police officers in Texas example: 20 employees - He contrasted a small-town department with a large urban agency to explain why controls must scale differently. California database abuse reports: 7,000 reported abuses last year - He cited this as evidence abuse is an industry-wide issue. Cities turning cameras back on: close to 20 cities - Langley said some cities reversed course after realizing they were less safe without Flock. Current customer churn: 1% - He said churn rose from zero to about 1% amid backlash. Drone speed: 60 miles per hour - He described drones as capable of reaching incidents quickly. Drone altitude example: 400 feet - He said drones can fly at about 400 feet. Drone zoom: 40x optical zoom - He used this to explain how officers can assess a scene remotely. Dallas-Fort Worth case response time: under a minute - A drone reached a reported armed subject at a gas station in under a minute. New Hampshire LPR retention: 3 minutes - He cited New Hampshire as having an extremely short retention rule. New York state retention: 21 days - He contrasted New York’s policy with New Hampshire’s.

Pivotal Quotes: "Data is a liability done asset." — Garrett Langley: Langley explained why retention limits matter and why Flock favors short defaults with local control. "We never want to be a company that defines suspicion." — Garrett Langley: He was discussing the limits of AI and predictive policing. "You don’t get to pick when you’re the victim of violence. You don’t get to pick." — Garrett Langley: He described why some former critics became advocates after personally needing Flock’s help.

Implications: Public-safety tech will face rising scrutiny. Winners will need strict defaults, auditability, transparent governance, and careful AI use. Communities will likely demand shorter retention, third-party oversight, and clear proof that tools improve safety without becoming surveillance infrastructure.