Document Category: Police

This motion seeks discovery concerning the reliability of colorimetric presumptive field drug tests, arguing that these tests are unreliable and prone to high false-positive rates. Drawing on empirical research and audits from jurisdictions nationwide, the motion explains that common legal substances frequently trigger positive results and that accuracy varies widely based on officer training, storage, and administration. It argues that information regarding test methodology, error rates, officer training, and agency policies constitutes Brady and Giglio material and is essential for effective assistance of counsel at the plea-bargaining stage and knowing pleas. Defenders can use this motion to compel early discovery.

This motion argues that Flock’s ALPR and Vehicle Fingerprint® systems are secretive, unregulated, and error-prone artificial intelligence tools whose reliability cannot be assessed without robust discovery into how they function, how they were developed and trained, and what their error rates are (pp. 4-7). It explains that machine-learning “hallucinations,” environmental conditions (weather, lighting, camera noise), and reliance on flawed databases and private HotLists undermine the accuracy of Flock outputs (pp. 4–7). Drawing on Maryland Rule 4-262, Brady/Giglio, and Crane v. Kentucky, the motion contends that Flock operates like an uncorroborated confidential informant or other similar technologies like ShotSpotter and Facial Recognition and that the accused is therefore constitutionally entitled to underlying data, logs, policies, audits, and technical documentation so they can confront and impeach Flock-derived evidence and present a complete defense (pp. 8–14).

This motion seeks to invalidate a warrant and suppress all resulting evidence where the warrant rests solely on Flock data. It explains that Flock’s machine-learning model generates conclusions based on a non-public AI model with an unknown error rate, is vulnerable to hallucinations, and is highly sensitive to weather, lighting, noise, and partial plate captures, with studies showing accuracy can fall to “near zero” under certain conditions (pp.1-6). Because Flock functions like an uncorroborated anonymous informant, the motion argues that Flock data cannot establish sufficient veracity for probable cause and lacks particularity (pp. 7-12). It further contends that omissions about Flock’s known flaws defeat the good-faith exception (pp. 12-15).

Relying on empirical studies from New York City, Chicago, Houston, and Dayton, Ohio, this motion explains that ShotSpotter alerts are unreliable because they rarely lead to discovery of gun-related crime or weapon use. It further argues that ShotSpotter alerts are unparticularized because they improperly equate assessments about a place with assessments about individuals in that place. Finally,  it contends that police should not be permitted to combine a ShotSpotter alert with other vague and unparticularized hunches—like the high-crime-area label—to establish reasonable suspicion, particularly given cited research about how ShotSpotter sensors are predominantly placed in communities of color and police are more likely to describe these same communities as high-crime areas regardless of actual crime rates.

This motion relies on new social science research showing that there is no serious risk of violence to police who conduct routine traffic stops to argue that courts should revisit the holding in Pennsylvania v. Mimms, 434 U.S. 106 (1977) (permitting officers to automatically order drivers out of the car during traffic stops) both under federal and state constitutional law. Instead of permitting officers to automatically order drivers out of cars during traffic stops, the motion argues that police should only be permitted to order drivers out of cars when they have reasonable suspicion that the driver poses a danger or is engaged in independent criminal activity – a standard that Hawaii, Massachusetts, and Vermont already use.  This motion can also be adapted to address cases in which police order passengers out of cars under Maryland v. Wilson, 519 U.S. 408 (1997).

This suppression motion argues that client’s statement—“I think it’d probably be a good idea for me to get an attorney”—should be understood as a clear invocation of the Fifth Amendment right to counsel. Relying on a recent empirical study, the motion argues that the invocation standard should be synonymous with a reasonable listener standard. Because ordinary American listeners overwhelmingly interpret conditional statements, questions, and hedges, like the one used by client, as clear invocations of the right to counsel, the motion argues a reasonable officer should have understood client’s statement to be a clear invocation of the right to counsel. In the alternative, the motion argues that the Court should follow the lead of other states and impose upon police a duty to clarify. Finally, the motion argues that because officers ignored client’s stated desire to speak with an attorney, the resulting statement was also involuntarily given.

This draft motion argues that courts should discount officer observations of “blading” when conducting reasonable suspicion analyses, because research shows that blading is a conclusory term that describes amorphous, innocuous, and contradictory behavior. This is particularly true when the term is applied to Black and brown community members who are trained to fear and avoid police. The motion asks the court to join others across the county in holding that “blading” is a generic label based on police hunches and is insufficient to establish reasonable suspicion of criminal activity.  These same arguments could be applied to reliance on “blading” as indicative of criminal activity for probable cause analyses.

This draft section of a motion to suppress details how judges’ and jurors’ assessments of the voluntariness of consent are likely to be impaired by a systematic psychosocial bias that makes them underappreciate the degree to which suspects feel pressure to comply with police search requests and overestimate the likelihood of freely-given consent. These tendencies persist even if police inform individuals that they have the right to refuse consent [p. 8]. This motion ultimately encourages courts to require reasonable suspicion before permitting requests for consent as NJ, RI, and CT already do [p. 8-10] or – at the very least – to take into account these psychosocial realities when analyzing the totality of the circumstances to determine whether consent was voluntarily given.  The research in this motion and the arguments raised could also be used to address the voluntariness of a decision to waive Miranda rights and give a statement to police.

This motion to suppress section argues that courts should discount officer statements that events took place in a “high-crime” area when conducting reasonable suspicion analyses. (The same data and arguments could also be used to discount “high-crime neighborhood” assertions in probable cause analyses.) The cited social science research demonstrates that (a) police are inconsistent about how they define high-crime areas (pp. 3-4); (b) police designations of high-crime areas are not accurate in defining actual crime rates or predictions for criminal behavior (p.4); and (c) police use high-crime area designations disproportionately against Black individuals (pp. 5-7).

This ten-page sample motion argues that judges should not consider a client’s record of prior police contacts, including arrests, that did not result in criminal convictions for purposes of sentencing because such records are (1) inherently unreliable/ambiguous and (2) likely to exacerbate existing racial disparities in the criminal legal system. The motion draws on national data and jurisdiction-specific case studies to show that people of color, particularly Black Americans, are stopped, searched, arrested, and charged at disproportionately high rates, not because of higher rates of crime commission but because of implicit bias in law enforcement. This sample motion also uses local data from Washtenaw County, Michigan, that defenders should replace with their own jurisdiction’s data when available, to demonstrate that these national trends are reflected in the defendant’s jurisdiction as well.  Data from this motion could also be useful to defenders drafting 4th Amendment suppression motions or making evidentiary arguments at trial to exclude evidence of prior police contacts.  To the extent that the court or the prosecutor bring up prior police contacts at pre-trial release hearings, this data could also be useful to argue for exclusion of prior contacts during bail/pretrial release assessments.

This motion relies on social science demonstrating the unreliability of witness attempts to identify people from surveillance videos/photographs and argues that due process and the evidence rules (Rules 602, 701, and 403) require exclusion of a police officer’s attempt to identify the defendant from a surveillance video.

Pages 2-4:  Discuss studies showing that humans are bad at matching people to images in photos/videos

Pages 4 –9:  Discuss studies showing that image quality (resolution, distance from subject, and moving versus still images), camera angle and viewpoint, lighting conditions at the time of the video or image capture, the presence or absence of obstructions to the camera’s view, and the size of the image captured all affect reliability

Pages 9 – 13: Discuss how situational factors including a lack of prior familiarity, cross-racial identification problems, and time delays between a prior exposure and the viewing of a surveillance photo/video all contribute to mistaken non-eyewitness identifications

Pages 15-16 – Discuss how these studies could also be used to (a) limit or prevent the prosecution from asking the judge/jury to compare a surveillance video/photo to the defendant, (b) obtain favorable expert testimony about the problems of non-eyewitness identification; (c) get the court to take judicial notice of these problems; (d) obtain favorable jury instructions about the problems with non-eyewitness identification testimony; (e) cross-examine non-eyewitnesses more effectively.

In this amicus brief, Fourth Amendment scholars draw on empirical research about the application of the good faith exception to the exclusionary rule to explain how overreliance on it has created perverse incentives for law enforcement and stunted the development of Fourth Amendment law – particularly in the context of digital data searches. Pages 4-5 explain the empirical study and its results.  Relying on this data, the brief argues for narrower interpretations of the scope of the good faith exception going forward.

This motion incorporates studies of implicit bias and racially discriminatory policing in San Diego. Studies referenced describe disparities in how police speak to drivers of different races as well as disparities in post-stop outcomes.

p. 13-59 overview studies of Shotspotter accuracy and racially biased implementation

Pages 1-4 discuss the history and development of the DRE program; pages 6-10 explain how the DRE test fails Rule of Evidence 702 because it does not assist the trier of fact to understand a fact in issue and the officers who testify about it are not qualified in the relevant field of knowledge; and pages 11-26 explain why the DRE protocol fails each of the five Daubert factors.

Cites studies demonstrating racial disparities in incidents of police violence, being stopped by police, and resulting perception of police

Based on six years of data of police stops in Philadelphia based on the smell of marijuana, this brief overviews how its use as a justification for a search has vastly increased the number of searches conducted, but less than 20% of the searches relying on the presence of marijuana actually discovered contraband. The brief argues that such discretionary opinions perpetuate implicit bias and result in racially biased outcomes.

This document was filed in support of a motion to compel discovery of the underlying source code, parameters, error rates, input data, results, reports, analyst, and confidence scores of the Facial Recognition program used in a criminal case. The expert explains the steps involved in a facial recognition search and areas where subjective human decisions — and cognitive bias — are likely to impact the outcome of the search.

Civil rights complaint about problems with facial recognition technology – pgs. 9-12 collect research about errors with when images are of low quality, angles are different, resolution is bad; pgs. 12-15 collect research showing facial recognition algorithms are racially biased; pgs. 15-16 collect data about jurisdictions that have banned use of facial recognition technology because of its flaws

Incorporating statistical evidence of racial disparities in police stops and police violence, this brief argues that “what constitutes “suspicious” or “abnormal” behavior for Fourth Amendment purposes must take into account the realities of racism and police violence experienced by communities of color” and that any totality of the circumstances analysis must consider “the fraught, frightening, and sometimes fatal treatment some communities receive from police.”

This brief challenges a police officer’s testimony opining that a shooter pictured in surveillance footage and the defendant were the same height.  The officer based his opinion on his visual observation of surveillance footage and measurements of the height of markings in the store. The Florida appellate court agreed that this was impermissible lay opinion testimony that invaded the province of the jury and required specialized expertise (https://caselaw.findlaw.com/court/fl-district-court-of-appeal/2076011.html).

Pages 17 – 23:  Discuss the science of photogrammetry – the process of discerning the size of objects in a photograph – and explain how estimates of an individual’s height in a photograph or video require expert calculations based on geometry, physics, and photogrammetric triangulation.  As a result, the brief argues that only a qualified expert can opine about the height of an individual in a photo or video, and a police officer’s opinion based on visual observation of the photos/video is unreliable, untethered from science, and inadmissible under both due process and evidentiary rules.

Argument that using in-court IDs as the only identification in case is a violation of the Due Process Clause because they create a substantial risk of misidentification. Social science studies cited throughout, specifically supporting reliability concerns (pgs. 10-12) and policy arguments (pgs. 13-15)

p. 10-12 discuss a study that found that people with normal smell identification abilities could only smell a 5-pound bag of marijuana in a trunk 13% of the time, and that 10% of the time people thought they smelled marijuana when there was none. p. 12-13 argues that an officer can impermissibly taint a search by suggesting that he smells marijuana, similar to tainting a lineup by suggesting the suspect to the witness.