Oversight & Accountability

Federal Court Intervention: When Oversight Fails, Courts Step In

Georgia's prisons have required federal court takeover twice in fifty years — a fact that renders every claim of adequate internal oversight structurally implausible. The first intervention came in *Guthrie v. Evans* (1972–1999), when federal courts assumed supervisory control of Georgia State Prison after finding conditions unconstitutional. That litigation lasted 27 years, longer than many of the sentences being served inside. The second intervention arrived in 2024, when a federal court imposed daily fines of $2,500 — $75,000 per month — on the Georgia Department of Corrections for 'flagrant' violations of a settlement agreement governing conditions in the Special Management Unit (SMU). (*Solitary Confinement & Restrictive Housing*) The fact that GDC was under an active consent decree and still required contempt sanctions demonstrates that court orders alone, without structural enforcement capacity, cannot substitute for functioning oversight.

The October 2024 Department of Justice investigation documented conditions that internal GDC oversight had failed to prevent or even formally acknowledge: 142 homicides between 2018 and 2023, staffing vacancy rates exceeding 50% system-wide with eight facilities above 70%, 27,425 weapons recovered in less than two years, and 12,483 contraband cellphones — all inside facilities that GDC certified as operating within policy. (*DOJ Investigation of Georgia Prisons*) Five homicides at four different prisons occurred in a single month in 2023. (*Prison Classification Systems & Violence*) The DOJ's findings did not represent new problems discovered from the outside; they represented problems GDC had documented internally and declined to escalate, correct, or publicly report.

The constitutional framework for this oversight failure is well-established. Under the Eighth Amendment and decades of Supreme Court precedent, states bear an affirmative obligation to protect incarcerated people from harm. Georgia's failure to staff its prisons, control violence, and provide constitutionally adequate conditions is not a resource problem alone — it is a governance problem. The $634 million in emergency corrections spending approved in 2025 (*Georgia's $600 Million Prison Spending Infusion*) arrived only after the DOJ investigation made the failure undeniable. Internal oversight mechanisms had existed throughout the preceding crisis and produced nothing.

The Accountability Gap in Mortality Reporting: What GDC Counts and What It Doesn't

The most direct evidence of GDC's oversight failure is the gap between what the agency reports and what independently verifiable evidence shows. In 2024, GDC reported 66 homicides. Georgia Prisoners' Speak independently tracked 100 homicide deaths in the same period — a discrepancy of 34 deaths. (*Who Is Responsible for Violence in Georgia's Prisons?*) This is not a rounding error or a definitional dispute. It is a 52% undercount of violent deaths in state custody. Deaths that GDC cannot or will not account for represent people whose families received no explanation, whose cases generated no investigation, and whose deaths produced no corrective action.

The national context makes Georgia's reporting failure even more damning. The Bureau of Justice Administration reported 5,674 deaths in custody nationally for fiscal year 2020, figures that researchers widely understand to be significant undercounts. (*Prison Mortality & Deaths in Custody*) Georgia's internal reporting problems exist within a national system that already systematically undercounts custodial mortality — meaning the documented 142 homicides between 2018 and 2023 (*DOJ Investigation*) are themselves likely a floor, not a ceiling. The drug overdose death trajectory tells the same story: from 2 deaths in 2018 to at least 49 between 2019 and 2022, with at least 5 additional confirmed deaths through mid-2023 (*Georgia Prison Drug Research*).

Field Drug Test Unreliability: A Structural Failure Entering Georgia's Prisons from the Front Door

The accountability failures that define Georgia's prison system do not begin at the prison gate. Many of them begin at the roadside, where law enforcement agencies across the state continue to use colorimetric field drug tests that are demonstrably, documentably unreliable — and where Georgia law makes those unreliable results uniquely dangerous.

Georgia is the only state in the United States where presumptive field drug test results remain admissible at trial for non-marijuana drug cases. In *Collins v. State* (278 Ga. App. 103, 2006), the Georgia Court of Appeals held that "positive field test results are alone sufficient to sustain a conviction for selling or possessing cocaine." In *Fortune v. State* (304 Ga. App. 121, 2010), admissibility was upheld under the Harper standard, which requires only that a technique has "reached a scientific stage of verifiable certainty" — a threshold the court applied without examining the tests' actual error rates. This legal framework exists despite the Department of Justice determining in 1978 that field drug tests "should not be used for evidential purposes," and despite the National Bureau of Standards cautioning as early as 1974 that the tests "should not be used as sole evidence for the identification of a narcotic or drug of abuse." Georgia courts have embedded as legally sufficient evidence a tool that federal scientific bodies rejected for that purpose more than four decades ago.

The error rates justify that rejection. While manufacturers have historically claimed error rates of approximately 4%, the Quattrone Center estimated actual false-positive rates range between 15% and 38%, depending on jurisdiction and substance. In New York City, the Department of Investigation submitted 71 items that had field-tested positive for fentanyl in city jails to an independent laboratory. Only 15% actually contained fentanyl — an 85% average false-positive rate. The NARK II tests manufactured by Sirchie — the same brand used in Georgia — produced a 91% false-positive rate in that testing. A Suffolk County Superior Court judge characterized NARK II kits as "arbitrary and unlawful guesswork" and their accuracy as "only marginally better than a coin-flip." In *Green v. Massachusetts Department of Correction* (2021), 38% of incoming prison mail that tested positive for synthetic cannabinoids contained no illegal drugs. The Colorado Department of Corrections found a 33% false-positive rate for its own colorimetric testing program — a finding that directly prompted legislative reform.

Georgia's own documented record is consistent with these figures. A 2018 FOX 5 Atlanta investigation confirmed 145 false positives from field tests statewide in a single year — 2017 — based on GBI Crime Lab records. A Savannah Police Department internal audit in 2017 found the portable drug test was wrong in 9 of 42 cases reviewed, a 21.4% error rate. The Quattrone Center estimates that approximately 961 Georgians are falsely arrested each year due to faulty field drug test results. The Georgia Innocence Project has identified invalid forensic evidence as a factor in 44% of its exoneration cases. Nationally, the National Registry of Exonerations documents that 531 of its 3,396 known exonerations involved wrongful drug arrests for substances that were not drugs.

The human cost of these errors is not abstract. In December 2016, Dasha Fincher of Monroe County, Georgia was arrested after cotton candy tested positive for methamphetamine in a Sirchie NARK II field test. Her bail was set at $1 million. She spent 94 days in jail. The deputy who arrested her, Cody Maples, admitted before a grand jury that he had no training on how to use field drug tests. Her federal lawsuit was dismissed — the court granted sovereign immunity to the deputies. Judge Tilman Self III wrote: "Without a doubt, Plaintiff should never have spent 94 days in jail. And while the Court certainly empathizes with her, it nonetheless must follow the requisite law." In October 2015, Ju'zema Goldring of Atlanta was arrested after stress ball sand tested positive for cocaine and spent nearly 6 months in Fulton County Jail. These are not outliers — they are the predictable output of a system that the Quattrone Center has called the use of colorimetric field testing "one of the largest, if not the largest, known contributing factor to wrongful arrests and convictions in the United States."

The racial dimension of this failure is not incidental. Black Americans experience erroneous drug arrests from field tests at a rate three times higher than white Americans on a per-capita basis. In Harris County, Texas — where at least 298 people were convicted of drug possession between 2004 and 2015 despite crime lab tests later finding no controlled substances — 60% of those wrongfully convicted were African American, in a city that is approximately 24% Black. The innocent in Harris County pleaded guilty an average of four days after arrest, well before any laboratory could return results. Ninety-three percent of those wrongfully convicted in field drug test cases examined by ProPublica received jail or prison sentences.

These coercive dynamics are structural, not incidental. Approximately 95% of criminal cases nationally are resolved through plea agreements. The Quattrone Center found that 89% of prosecutors accept guilty pleas without any confirmatory laboratory testing, and 67% of drug labs report they are not asked to review samples when cases are resolved by plea. Over 80% of prosecutors acknowledged it is "extremely unlikely" that seized drug evidence will ever be analyzed once a plea deal is reached. The Colorado working group convened to study the issue stated plainly: "people regularly plead guilty to drug possession offenses absent laboratory confirmation because they cannot afford to remain in custody awaiting a laboratory test." The same working group concluded that "in the absence of laboratory confirmatory testing, the incidence of false positives is largely invisible." In Georgia, that invisibility is compounded by the GBI evidence backlog, which reached 36,194 items in 2019, with 19,112 in the chemistry and drug testing section alone. Macon District Attorney David Cooke reported 7- to 8-month waits for results in that period. In 2017, at least three people had already pleaded guilty before GBI lab results returned showing no controlled substances. In 2024, the GBI lab received approximately 103,000 testing requests and reported back approximately 105,000 — the first time output exceeded input in several years, reducing the backlog by 11%. Progress is real; the structural incentive to plead before results arrive is not resolved by it.

Georgia compounds this nationally shared problem through a feature unique to itself: in Georgia, possession of any amount of a Schedule I or II controlled substance — cocaine, heroin, fentanyl, methamphetamine — is classified as a felony under O.C.G.A. Section 16-13-30, carrying penalties of 2 to 15 years. Every false positive in Georgia is a false felony. Despite this, 68% of Georgia police agencies responding to a 2024 FOX 5 Atlanta survey still use roadside drug tests, and only 26% had a policy requiring additional evidence before making an arrest based on field test results. There is no mandatory training in Georgia for officers on how to use field drug tests.

Field Drug Tests Inside Georgia Prisons: Discipline Without Due Process

The false-positive problem does not stop at the prison gate for people who are wrongfully convicted. It continues inside Georgia's prisons in a parallel, less-scrutinized context: the use of colorimetric field tests to discipline incarcerated people for alleged drug possession without any confirmatory laboratory testing. In New York, the State Inspector General found that 2,000 inmates had been wrongly punished based on unverified field test results, including placement in solitary confinement, suspended visitation, and loss of earned good-time credits. In *Green v. Massachusetts Department of Correction* (2021), the 38% false-positive rate on incoming prison mail directly resulted in disciplinary sanctions against people who had received no contraband. Nebraska passed LB 519, a narrower measure allowing prison inmates who receive false positives to request confirmatory retesting before disciplinary action is taken — an acknowledgment that the problem is real and that the default practice of acting on unconfirmed results causes harm. Georgia has no equivalent protection.

The Reform Gap: What Colorado Did and What Georgia Has Not

On March 26, 2026, Governor Jared Polis signed Colorado House Bill 26-1020, making Colorado the first state to legislatively address colorimetric field drug testing. The bill passed 65-0 in the House and 33-0 in the Senate, with an estimated fiscal impact of $0 in new appropriations — summons procedures slightly increase court workload but reduce jail booking costs. The statute prohibits arrests for Level 1 drug misdemeanor possession or municipal drug possession charges when a colorimetric field drug test is the sole basis for the charge. Officers must instead issue a summons and submit evidence for laboratory testing. Before accepting any guilty plea for Level 4 drug felony possession or lower where a colorimetric test was used, courts must advise defendants that colorimetric field drug tests are subject to false positives and that laboratory confirmation is available.

Colorado's action followed years of documented internal failure. The working group created by the predecessor bill, HB 25-1183, surveyed prosecutors in all 23 of Colorado's judicial districts; only four responded, but those four unanimously confirmed that lab tests are only conducted when cases go to trial — meaning the overwhelming majority of convictions rest on field test results alone. The Las Vegas Metropolitan Police Department's own crime lab director had wanted to abandon field drug test kits since 2010; the department continued using them for years. Jacksonville, Florida's Duval County Sheriff's Office stopped using field tests entirely in September 2024 after discovering multiple over-the-counter cold medications triggered false cocaine positives.

The American Legislative Exchange Council finalized a model Colorimetric Presumptive Field Drug Test Limitations Act in January 2026, more expansive than Colorado's law, barring colorimetric results from being used for probable cause, arrest, charging, conviction, or sentencing. California's SB 912 — the ROAD Testing Act — would have prohibited colorimetric test results from being used for probable cause or arrest but died in committee due to a two-thirds supermajority requirement. North Carolina's HB 868 has been stalled in committee since April 2025.

Georgia has taken no comparable legislative action. A Georgia reform bill would need to address five structural elements: mandating summons in lieu of arrest for simple possession based solely on field tests; requiring court advisements before guilty pleas; restricting admissibility of unconfirmed field test results at trial; mandating confirmatory testing before felony charges proceed; and establishing training requirements for officers who use field tests. None of these requirements currently exist. Governor Kemp signed the Wrongful Conviction and Incarceration Compensation Act in 2025, providing $75,000 per year to exonerees, and stated that "Georgia values justice even when it means admitting error." The field drug test framework Georgia currently maintains makes wrongful convictions not an aberration to be compensated after the fact, but a predictable and preventable output of deliberate policy choices that remain unchanged.