ShotSpotter Acoustic Gunshot Detection
Chicago ran ShotSpotter acoustic sensors across 12 of its 22 police districts, more than 100 square miles of the South and West Sides. The city's Inspector General found in 2021 that 9.1% of police responses to alerts turned up evidence of a gun crime. Mayor Brandon Johnson let the contract lapse; CPD stopped using the system on 23 September 2024. No replacement has been procured since.
Completed
📅 2012–2024
✔ Editor-reviewed
📊 Impact
The Chicago Office of Inspector General reviewed every alert from 1 January 2020 to 31 May 2021: of 50,176 alerts confirmed as probable gunshots, 41,830 recorded a disposition, and 4,556 of those found evidence of a gun-related offence — 9.1% of police responses. A documented investigatory stop followed 2.1%. A synthetic-control evaluation of Chicago and Kansas City funded by the National Institute of Justice found that the software detected gunfire as advertised, but that gun violence did not fall in either city after ShotSpotter arrived and shootings were no more likely to be solved.
🎓 Lesson
Accuracy was never the question Chicago could not answer. The city could hear the shots; what its own watchdog could not find was what the police response produced.
Evidence📊 Independent
Scale🏙️ City-wide
Cost≈$53M total contract value (initially $33M for 2018-2021)
FundingCity of Chicago (Chicago Police Department budget)
Sources
↗ Chicago Office of Inspector General — The Chicago Police Department's Use of ShotSpotter Technology (24 Aug 2021)The load-bearing document. Verbatim: 'A total of 50,176 ShotSpotter alerts were confirmed as probable gunshots'; 'a total of 4,556 indicate that evidence of a gun-related criminal offense was found, representing 9.1% of CPD responses'; 'at least one investigatory stop is documented under a matching event number in 2.1% of all CPD responses.' Also dates the contract: '$33 million, three-year contract with ShotSpotter began on August 20, 2018', extended in December 2020 to 19 August 2023. The OIG states its own data were incomplete and irreconcilable. Caution for a future check: the report PDF did not yield machine-readable text in this review, so these figures are quoted from the OIG's release page rather than the PDF body.
↗ Chicago Office of Inspector General — full report PDF (August 2021)The report itself, for readers who want the methodology and the four recommendations. 985 KB.
↗ National Institute of Justice — Piza et al., The Impact of Gunshot Detection Technology on Gun Violence in Kansas City and Chicago: A Multi-Pronged Synthetic Control EvaluationThe independent quantitative evaluation, and the reason the entry does not simply repeat advocacy figures. Chicago sample: 85,572 incidents, 6 February 2017 to the end of 2019. Note for anyone reusing it: the frequently quoted procedural gains — alerts arriving 93 seconds before 911 calls, locations about 300 feet closer — are the KANSAS CITY results in this study and are not Chicago findings.
↗ Northeastern University — ShotSpotter improves detection and response to gunfire, but doesn't reduce crime (23 May 2024)Plain-language account of the same evaluation by its lead author, Eric Piza. Carrying sentence: 'Gun violence did not reduce in either (Kansas City or Chicago) following the introduction of ShotSpotter, and shootings were not anymore likely to be solved in either city.' Piza's own site states the project also produced peer-reviewed journal articles; those were not individually verified here, which is why this entry is graded independent rather than peer-reviewed.
↗ MacArthur Justice Center (Northwestern Pritzker School of Law) — ShotSpotter Generated Over 40,000 Dead-End Police Deployments in Chicago in 21 Months (May 2021)Litigation-side research, published three months before the OIG audit and reaching a similar shape from different data: 1 July 2019 to 14 April 2021, 89% of deployments turned up no gun-related crime and 86% no crime report at all. Labelled here as advocacy research, not an audit — the entry's headline figure is the OIG's, not this one.
↗ SoundThinking — the vendor's rebuttals to the OIG audit and to the MacArthur Justice Center reportThe other side, kept in deliberately. SoundThinking argues the OIG measured the wrong thing (dispositions of police responses rather than detection accuracy) and published a separate point-by-point response to the MJC study at soundthinking.com/blog/soundthinkings-response-to-the-macarthur-justice-center-report/, plus a commissioned Edgeworth Economics critique of its methodology. The company also renamed itself from ShotSpotter, Inc. on 10 April 2023; the product kept the old name.
↗ ABC7 Chicago — Chicago ShotSpotter contract ends Sunday, city to begin dismantling technology Monday (22 Sep 2024)The end, with the city's own wording: 'Effective Monday, September 23, 2024, at 12:01 am, the Chicago Police Department will no longer be utilizing ShotSpotter technology', followed by a 60-day decommissioning. Mayor Johnson's stated reason: 'The ultimate goal is to deploy resources on the most effective strategies and tactics proven to accelerate the current downward trend in violent crime. We have to explore better options that save more lives.' The city's own statement page (chicago.gov, February 2024) carries the same text but returns 403 to automated fetching.
↗ WTTW — City Council Votes 33-14 to Again Rebuke Mayor Brandon Johnson's Decision to Scrap ShotSpotter (18 Sep 2024)The fight over the ending, which is not settled. The council voted 33-14 to hand contracting authority to Superintendent Larry Snelling, who supports the technology; Johnson called the ordinance illegal, Snelling did not sign over the objection of the mayor who appointed him, and the system went dark four days later.
↗ Chicago Sun-Times — City Council members want Johnson to tackle replacing ShotSpotter (12 May 2026)Currency check, and the reason this entry says nothing has replaced it. Twenty months after shutdown: 15 firms answered an RFI, 9 answered the RFP including SoundThinking, $13.9M sits in two budgets, no contract is awarded. Chief Procurement Officer Sharla Roberts: 'An RFP of this size, complexity and nature takes 18 to 24 months and sometimes longer to award.' A resolution filed 5 May 2026 with 35 council signatures accuses the mayor of obstruction; hearing set for 3 June 2026. A free six-month muzzle-flash pilot from Shooter Detection Systems was offered in the 9th District in January 2025; no source confirms it went live.
↗ Chicago Sun-Times — Chicago settles ShotSpotter lawsuit, considers bringing gunshot-detection system back (15 Aug 2025)The legal counterpart to the OIG's finding that the technology changed police behaviour: Chicago paid $90,000 to settle a 2022 federal suit by Daniel Ortiz and Derick Scruggs and conceded that a ShotSpotter alert alone does not justify stopping or frisking someone near the alert. Separately, the city agreed in March 2026 to pay $500,000 to Michael Williams, charged with murder in 2020 largely on an alert and held nearly a year before prosecutors dropped the case in July 2021 for lack of evidence.
↗ WTTW — Steep Drop in Homicides Continued for Full Year After ShotSpotter Was Removed: Analysis (29 Dec 2025)What happened after, reported here as correlation and nothing more. Police beats that had had sensors saw roughly a 32% fall in homicides between 1 January and 22 September 2025, but homicides fell about 30% citywide in 2025 to a 60-year low, including where sensors never existed. A University of Chicago analysis reported on 27 May 2026 found faster responses to top-priority 911 calls in former ShotSpotter beats, but it excluded gunfire calls from the data — a limitation critics reasonably press. Both WTTW pages return 403 to automated fetching; figures here come from search-result text and should be re-read before any figure is added to the entry body.
📎 Cite this project
1001 Smart Cities (2026). “ShotSpotter Acoustic Gunshot Detection” — Chicago, United States. The Smart City Atlas. https://1001smartcities.org/projects/chicago-shotspotter/ (last verified 2026-07-26). Data: CC BY 4.0.
The underlying data is free to reuse with attribution — see the open data page.
🔗 Related projects
🇺🇸 Surveillance Technology Ordinance & Facial Recognition Ban · San Francisco
San Francisco requires a published impact report and a Board vote before a department may acquire a technology like this.
🇺🇸 Operation LASER & PredPol — Predictive Policing · Los Angeles
Both ended after their own inspector general looked: LA could measure no crime effect; Chicago found evidence in 9.1% of responses.
🇺🇸 Array of Things · Chicago
Chicago's other sensor network, same years: the Array published air and noise readings openly; these microphones fed police dispatch.
🇮🇳 Citywide CCTV Rollout and Delhi Police Facial Recognition · New Delhi
Delhi's police told a court their matching ran at 1-2% and kept the system; Chicago measured 9.1% and let the contract lapse.
→ Open this project in the interactive atlas · → 📡 Data & IoT category · 🚩 Report an error