ShotSpotter / SoundThinking gunshot detection (Chicago Police Department)

entry · p2026-09-06-262 · validated

The total number of people stopped in these events is 1,740

tech: shotspotter
passes the definition as: a machine-learning classifier fit to recorded impulsive sounds filters candidate gunshots; the learned part ends there — a human acoustic reviewer, not the model, decides whether to publish the alert, and the stop is the officer's
from: 2020-01
region: US
domain: agency
side: loss
people: 1,740 people stopped (2021-05)
effect: 1,740 people stopped (2021-05) — people stopped by police under the 1,056 alert event numbers that matched an investigatory stop report over 17 months; 825 of those alert events recorded at least one pat down and 244 at least one arrest
reversible: yes
who holds the switch: owner:Chicago Police Department
attribution: named
source kind: primary
the statistic: Chicago OIG, August 2021: of 50,176 ShotSpotter alerts confirmed and dispatched between 1 January 2020 and 31 May 2021, 9.1% of police responses produced evidence of a gun-related offence and only 2.1% matched a documented investigatory stop. 1,740 people were stopped under those 1,056 alerts; 152 of the alert events recorded a gun recovered. Chicago switched the system off on 23 September 2024.
source: https://igchicago.org/wp-content/uploads/2021/08/Chicago-Police-Departments-Use-of-ShotSpotter-Technology.pdf
proposed by: researcher (opus), brief 2026-09-06, us-surveillance
received: 2026-09-06T14:27:54-04:00
harvest report: impacts/harvest/2026-09-06-us-surveillance.md

Rulings, in order

2026-09-06T14:27:55-04:00 · validatorpass — quote found on the fetched page (exact match; HTTP 200 application/pdf)
2026-09-06T14:28:26-04:00 · sessionseverity — Stopped and often patted down on a gunshot alert that a human reviewer published from the classifier's candidates: a decision about you, reversible, with recourse — ordinary. The learned part ends at the candidate list, so the stop is scored at the lowest row. Reach on the 1,740 people stopped (Chicago OIG, 2020–21).

The proposal as it arrived (JSON)

{
 "id": "p2026-09-06-262",
 "received": "2026-09-06T14:27:54-04:00",
 "proposer": "researcher (opus), brief 2026-09-06, us-surveillance",
 "kind": "entry",
 "sha": "548af03c28bc014c",
 "data": {
  "kind": "entry",
  "tech": "shotspotter",
  "tech_name": "ShotSpotter / SoundThinking gunshot detection (Chicago Police Department)",
  "definition": "a machine-learning classifier fit to recorded impulsive sounds filters candidate gunshots; the learned part ends there — a human acoustic reviewer, not the model, decides whether to publish the alert, and the stop is the officer's",
  "region": "US",
  "domain": "agency",
  "side": "loss",
  "from": "2020-01",
  "people": {
   "n": 1740,
   "unit": "people stopped",
   "date": "2021-05"
  },
  "effect": {
   "n": 1740,
   "unit": "people stopped",
   "date": "2021-05",
   "what": "people stopped by police under the 1,056 alert event numbers that matched an investigatory stop report over 17 months; 825 of those alert events recorded at least one pat down and 244 at least one arrest"
  },
  "reversible": true,
  "switch": "owner:Chicago Police Department",
  "statistic": "Chicago OIG, August 2021: of 50,176 ShotSpotter alerts confirmed and dispatched between 1 January 2020 and 31 May 2021, 9.1% of police responses produced evidence of a gun-related offence and only 2.1% matched a documented investigatory stop. 1,740 people were stopped under those 1,056 alerts; 152 of the alert events recorded a gun recovered. Chicago switched the system off on 23 September 2024.",
  "source": "https://igchicago.org/wp-content/uploads/2021/08/Chicago-Police-Departments-Use-of-ShotSpotter-Technology.pdf",
  "quote": "The total number of people stopped in these events is 1,740",
  "attribution": "named",
  "source_kind": "primary"
 },
 "from_file": "impacts/harvest/2026-09-06-us-surveillance.md"
}