ShotSpotter (Again) Spotted Altering Shots (And Spots) To Better Serve Police Narratives

(Mis)Uses of Technology

from the so-scientific-it-can-be-remodeled-on-the-fly! dept

Dozens of cities around the nation are relying on early warning tech to help their law enforcement get out ahead of crime. (Well, get out slightly behind crime, to be accurate…) Microphones and sensors placed in strategic locations around cities pick up loud noises and pass this information on to police departments so they can scramble cops to the spotted shot.

That’s what ShotSpotter does. How well it does it is still up for debate. SpotShotter detects loud noises that might be gunshots, runs everything through some coding, and makes a determination. Maybe the sound is just a backfire or someone setting off fireworks. Or maybe it’s actually gunfire.

This is what those in the cop business call “actionable intel.” The problem is that ShotSpotter isn’t as accurate as cops think it is. On the plus side, ShotSpotter appears to be every bit as malleable as cops want it to be.

A few years ago, we covered the story of a man who sued Rochester police officers after they used ShotSpotter data (after the fact) to justify shooting him. A lot of things in the narrative didn’t add up.

Rochester resident Silvon Simmons was headed home from a convenience store around 9 pm when a cop car cut him off. The officer hit Simmons with his spotlight and then opened fire when Simmons ran away from the car. At no time did the officer identify himself as an officer.

The Rochester PD attempted to justify this use of deadly force when faced with no deadly threat. Officers said they found a gun in a yard, but it was several houses away from where Simmons was stopped and in the opposite direction of where he had run when confronted by the police officer. The officers then said Simmons had fired on them, prompting them to fire back. Simmons’ DNA and fingerprints were not on the recovered gun.

So, the PD turned to ShotSpotter. What was originally determined to be “helicopter noise” by a ShotSpotter mic near the scene was changed to “multiple gunshots” at an officer’s request. Unfortunately for the officer, ShotSpotter’s count of shots fired was too few, as it only included the shots fired by officers. Another request resulted in ShotSpotter finding another gunshot in its recordings — the one supposedly fired by Simmons at officers from the gun he never had in his possession. That was good enough to get him charged with aggravated assault on a police officer. He spent a year in jail before being acquitted on all charges.

ShotSpotter and its law enforcement partners are doing it again. They’re altering evidence to fit narratives, as Todd Feathers reports for Motherboard.

On May 31 last year, 25-year-old Safarain Herring was shot in the head and dropped off at St. Bernard Hospital in Chicago by a man named Michael Williams. He died two days later.

Chicago police eventually arrested the 64-year-old Williams and charged him with murder (Williams maintains that Herring was hit in a drive-by shooting). A key piece of evidence in the case is video surveillance footage showing Williams’ car stopped on the 6300 block of South Stony Island Avenue at 11:46 p.m.—the time and location where police say they know Herring was shot.

The key connective evidence in this case comes from ShotSpotter. But since ShotSpotter didn’t originally deliver the evidence officers wanted, the evidence apparently had to change. The night in question, 19 ShotSpotter devices “detected a percussive” sound, determining the originating location of this noise to be 5700 Lake Shore Drive. That’s about a mile from the address where Williams’ vehicle was caught on camera.

Obviously, evidence putting a potential gunshot a mile away from where the murder was supposedly committed wouldn’t do. Alterations were made.

[A]fter the 11:46 p.m. alert came in, a ShotSpotter analyst manually overrode the algorithms and “reclassified” the sound as a gunshot. Then, months later and after “post-processing,” another ShotSpotter analyst changed the alert’s coordinates to a location on South Stony Island Drive near where Williams’ car was seen on camera.

It’s one thing to review an alert and flag it as a potential gunshot (rather than the original determination that this might have been a firework). It’s quite another to move the location of the detected noise to exactly where it does cops and prosecutors the most good.

This discovery has resulted in Williams and his lawyer raising a challenge to the evidence.

“Through this human-involved method, the ShotSpotter output in this case was dramatically transformed from data that did not support criminal charges of any kind to data that now forms the centerpiece of the prosecution’s murder case against Mr. Williams,” the public defender wrote in the motion.

The motion asks for the court to examine and rule on the evidence and ShotSpotter’s methodology to determine whether they’re scientifically sound. Unfortunately, we won’t be seeing any in-depth examination of ShotSpotter or law enforcement’s contribution to the shifting ShotSpotter narrative.

Rather than defend ShotSpotter’s technology and its employees’ actions in a Frye hearing, the prosecutors withdrew all ShotSpotter evidence against Williams.

It would be presumptive to call this an implicit admission of guilt. But I think we’ve earned that indulgence. If this were legit evidence — something both ShotSpotter and investigators could justify with thorough explanations and examination of ShotSpotter’s shot spotting tech — you’d think it wouldn’t have been dropped as soon as it was challenged. But it appears prosecutors would rather let an accused murder walk than discuss their tactics and their partner in (fighting) crime in open court.

Filed Under: law enforcement, police, shotspotter