An almost too-specific “gunshot detection technology grant” from the Florida Department of Law Enforcement covered the $280,000 price tag to monitor audio in a four-square-mile radius in East Tampa.
The motion passed 5-1 last month with councilmember Gwen Henderson, whose district includes ShotSpotter’s radius, absent at the vote.
Council member Lynn Hurtak, the lone “no” vote, voiced concerns and cited some 40,000 ShotSpotter alerts in Chicago, resulting in dead-ends for police.
A May 2021 study by the MacArthur Justice Center (MJC) on Chicago’s use of the software called ShotSpotter’s much-touted 97% accuracy rate into question. MJC found that 89% of calls turned up no gun-related crime, and 86% turned up no crime at all.
“Just because we have a grant doesn’t mean we have to take it,” Hurtak said at the meeting before voting against the resolution.
In response to the MJC study, ShotSpotter commissioned Edgeworth Analytics to review the MJC study independently. In two independent reviews commissioned by ShotSpotter, Edgeworth confirmed ShotSpotter’s 97% accuracy rate and refuted the MJC study.
Another review of ShotSpotter’s accuracy and Edgeworth’s analysis pointed out that Edgeworth determined accuracy primarily through customer satisfaction—reports from police clients using the system.
Alex Marthews, the founder of privacy advocacy group Digital Fourth, said a review of the Cambridge Police Department’s ShotSpotter usage showed that two-thirds of the time, officers could not verify a ShotSpotter report had actually heard gunfire. That could be because a shooting did take place and police could not find physical evidence, but also because the noise was inaccurately identified in the first place, Marthews said.
Jacob Wourms, research and campaign manager with Campaign Zero’s Cancel ShotSpotter project, told Creative Loafing Tampa Bay that ShotSpotter has “never proven its effectiveness.” His group’s primary goal is “public safety beyond policing.”
“Even when there’s a shooter, innocent people are often wrongly arrested,” Wourms told CL. “If there’s no shooter, people just going about their day can be arrested and brought into the criminal legal system because of this, too.”
Wourms said SoundThinking (the company formerly known as ShotSpotter) has never really submitted its software or findings for independent validation.
A spokesperson for SoundThinking quoted the Edgeworth audit in response, adding that “a live-fire test in Pittsburgh, PA, showing that the system detected 100% of incidents and located 96.9% of incidents within a 25-meter benchmark criterion.”
The spokesperson also sent along a study from Southern Illinois University Edwardsville which showed a 24% decrease in aggravated assaults in Winston Salem, N.C., after implementing ShotSpotter. Another study by the NYU Policing Project saw a 30% decrease in gun-related assaults.
According to a Nov. 8 memo from Tampa Police Chief Lee Bercaw, there have been 475 ShotSpotter alerts year-to-date (through Sept. 30, 2023), and for those activations, 13 arrests were attributed to ShotSpotter. Bercaw said that’s a 32% decrease in the 19 arrests made using ShotSpotter in 2022 but advocated for continuing its use.
“The Tampa Police Department believes that ShotSpotter is an effective tool…,” Bercaw wrote. “If ShotSpotter has led to the saving of one life or the solving of one crime, it has paid for itself.”
Hurtak noted that the $280,000 for ShotSpotter was initially built into the budget the council had later rejected. That’s why TPD sought a grant, in this case from Republican Sen. Jay Collins, whose district includes the City of Tampa.
Collins’ office has not responded to requests for comment.
“I just believe that three more police officers is a better way to spend our money,” Hurtak told CL on the phone recently.
Related
TPD’s controversial gunshot detection program unfairly targets East Tampa, councilmen say
ShotSpotter boasts over 160 cities using its software across the United States. However, over 30 cities have canceled or rejected contracts with ShotSpotter in recent years. Still, the company boasts a 99% renewal rate for subscribers.
San Antonio’s police chief pushed the city to end its contract with ShotSpotter in 2017 and said, “We’re going to use that money to provide more community engagement, which ShotSpotter can’t provide.”
SoundThinking told CL that “potential gunfire incidents are reviewed and validated by highly trained acoustic experts who analyze the audio and visual soundwaves to determine if they match the typical pattern of gunfire. After this secondary review, gunfire incidents are published to the police, and unrelated sounds are dismissed.”
But Wourms said there’s a question about how trained those reviewing ShotSpotter’s activations actually are.
“There’s a misconception that the people reviewing the audio are highly trained acoustic professionals,” Wourms said. “The reality is that these people are everyday folks like you and I. ShotSpotter’s job posting asks for a minimum of one year of professional experience, preferably in a call center. Most people reading your article would be qualified for this position.”
In Cleveland, where ShotSpotter has been in place since 2020, data published a day after Tampa’s vote showed 10 out of 18 people arrested for violent crimes following a ShotSpotter alert were ultimately found not guilty, or charges were dismissed.
“ShotSpotter manufactures probable cause, and we think that’s a bad thing,” Wourms added.
The Hillsborough County Sheriff’s Office (HCSO) decided against renewing its contract with ShotSpotter after 2019. Documents obtained through public records requests by CL show the HCSO agreement with ShotSpotter, signed under former Sheriff David Gee from 2016-2019, cost $800,000. On June 16, 2018, HCSO’s chief financial officer, Christina Porter, sent a certified letter to ShotSpotter saying the county “will not be renewing the ShotSpotter service agreement.”
That was after the department came under the leadership of Sheriff Chad Chronister. CL contacted HCSO for more information about why the contract wasn’t renewed but only received the above documents in response.
Beginning in 2019, Tampa’s ShotSpotter contract was grant-funded over the first three years. On Sept. 1, 2022, the City Council unanimously approved $280,000 in funds for another year-long contract. At that meeting, Deputy Chief Calvin Johnson, TPD’s Deputy Chief of Community Outreach and professional standards, told the council that the contract was for services already rendered. He also defended the software’s use for police.
Related
Tampa City Council will vote on funding for controversial gunshot detection program
“This is the third year in a three-year contract, and our violent crime rate has gone up more than other cities in the United States,” Carlson said to Johnson in response.
At that meeting, Carlson also asked Johnson about retired Tampa Police Captain Paul Lusczynski, who was working for TPD when ShotSpotter was implemented in 2019. Lusczynski retired from TPD in January 2022 and reportedly began working for ShotSpotter as a customer success director (it’s also listed on his LinkedIn).
In February of this year, Portland’s City Auditor investigated ShotSpotter for possible violations of lobbying codes. The investigation came just four days after the city closed its call for proposals to run a citywide gunshot detection pilot program. Oregon Public Broadcasting reported that emails showed Luscznski helped “arrange for Police Bureau representatives and members of the Focused Intervention Team oversight group to meet with Tampa police and see how the agency uses ShotSpotter.”
The visit was to coincide with a gun violence reduction conference in Tampa, and Lusczynski told the Portland delegation he’d “facilitate the introductions” to Tampa.
Earlier this month, at a Nov. 14 town hall following the deadly Oct. 29th shooting in Ybor City, Bercaw referred to a gun violence reduction summit that took place in Tampa in January of this year, just weeks before Portland’s investigation was launched.
“In January, we hosted a violent crime summit with all major cities,” Bercaw told attendees at the town hall. “They all came to see what the best practices were for violent crime…we’re still a model city.”
CL asked TPD if they knew of the Portland investigation and if Bercaw met with anyone from Portland during the summit to show them how Tampa was using ShotSpotter.
“Chief Bercaw did not meet with representatives from Portland In January,” Joneé Lewis, TPD’s director of communications and public relations, told CL via email.
Ultimately, Portland’s City Auditor found insufficient evidence that the company violated city regulations. But Portland decided against funding gunshot detection software after the investigation. This isn’t the first time questions have been raised about the ethics of how ShotSpotter is being sold to cities. In 2019, Oakland’s Public Ethics Commission fined the company $5,000 for violating the city’s Lobbyist Registration Act and Oakland Campaign Finance Reform Act.
“What we hear from some police is that we want to have tools in our toolbox, and ShotSpotter is just one tool,” Wourms told CL. “What we try to tell people is, if the tool doesn’t work, do you want it in your toolbox?”
In July this year, the Houston Chronicle reported that 911 call times increased from 60 minutes to 100 minutes after implementing ShotSpotter. Sound Thinking refuted the story.
“No matter what other call you’re on if (a ShotSpotter alert) drops, you go,” said Douglas Griffith, president of the Houston Police Officers’ Union. “You’re not going to that burglary report or that theft report. You’re now going to ShotSpotter.”
TPD would not say whether or not ShotSpotter picked up on the Oct. 29th Ybor City mass shooting and declined to explain how ShotSpotter activations work with 911 calls.
“Specific information regarding investigative techniques is exempt from public release,” Lewis, TPD’s director of communication and public affairs, wrote.
Lewis responded to CL’s additional request, pointing to the specific part of Florida’s broad public records laws that supported that claim.
“You inquired about a specific location for ShotSpotter. The reason we are only publicly releasing that the ShotSpotter system covers approximately a 4 square mile area in East Tampa is that specific information regarding investigative techniques is exempt from public release per 119.071 (2)(d). The memo presented to city council that was shared with you provided additional context about general coverage area,” Lewis wrote in an email.
SoundThinking stated that releasing that information is “at the sole discretion of the Tampa Police Department.”
A rep for Tampa’s police union also didn’t respond; Adam Smith, Communications Director for the City of Tampa, forwarded CL’s questions to TPD and offered Bercaw’s memo as a response.
If ShotSpotter is another tool in law enforcement’s toolbox, the company’s available tools are expanding. In September, Wired reported that Sound Thinking is buying parts of Geolitica, which created PredPol, a controversial predictive-policing technology.
“SoundThinking recently announced an agreement with Geolitica (formerly PredPol) to acquire its patents and select pieces of its intellectual property,” SoundThinking told CL. “To be clear – SoundThinking did not buy Geolitica. Rather, we came to an agreement to purchase specific assets from a company that is ceasing operations at the end of 2023.”
According to Wired, SoundThinking hired Geolitica’s engineering team and is taking over its existing customer base.
That makes SoundThinking’s acquisition of parts of Geolitica significant, Andrew Ferguson, an American University law professor and author of “The Rise of Big Data Policing,” told Wired.
“We are in a consolidation moment with big police tech companies getting bigger, and this move is one step in that process.”
A 2021 investigation by the Markup and Gizmodo (RIP) of PredPol’s data found that PredPol, which became Geolitica in March 2021, regularly targeted Black and Latino neighborhoods, including areas where residents qualify for federally free or reduced school lunches.
“What’s concerning with predictive policing is this idea that you can predict future crime based on historic crime,” Wourms said. “If you use ShotSpotter, a tool that is often wrong, and use it to try and predict future crime, they’ll be determining police patrols based on bad data.”
In a city like Tampa, desperate to curb a recent rise in violent crime, law enforcement, and leadership might look at other tools more controversial than ShotSpotter offered by companies like SoundThinking.
Subscribe to Creative Loafing newsletters.
Follow us: Apple News | Google News | NewsBreak | Reddit | Instagram | Facebook | Twitter | Or sign up for our RSS Feed
This article appears in Nov 30 – Dec 6, 2023.