End the Warehouse: Prison Transformation Plan

The Georgia Department of Corrections does not rehabilitate. It warehouses. This is the central premise of GPS's "End the Warehouse" platform — that a system holding nearly 50,000 people and consuming roughly $1.79 billion in state appropriations annually has, by design and by neglect, abandoned the rehabilitative mission its statutes still nominally claim. The evidence for that abandonment is not theoretical. It lives in the words of the people inside, who describe in firsthand testimony to GPS's Tell My Story project a daily reality of unsupervised dorms, withheld education, weaponized parole denials, and a culture in which compliance is punished and survival demands the opposite of the "rehabilitation" the agency markets to legislators. The sections below trace that reality through testimony, budget data, population figures, and policy — and lay out what transformation would actually require.

The Warehouse Is a Choice, Not an Accident

Georgia's prison spend is not small and not shrinking. According to the Governor's Budget Report, Amended FY 2026 and FY 2027, GDC drew approximately $1.42 billion in State General Funds in FY 2024, $1.82 billion in FY 2025, and $1.78 billion in the amended FY 2026 budget. HB 974 (FY 2027G), the Senate Appropriations Committee Substitute, approves roughly $1.76 billion in State General Funds and a total public-funds figure of $1.79 billion for FY 2027. Against that spend, GDC's own weekly population snapshots show the agency holding 49,952 people as of May 15, 2026, with a backlog of 2,530 awaiting transfer from county jails — a population that has hovered within a few hundred of 50,000 every Friday across the spring of 2026. The agency is therefore spending roughly $35,000 per incarcerated person per year, while the testimony collected by GPS describes facilities with empty security bubbles, dorms without supervision for hours at a time, and incarcerated people calling their mothers to get help during medical emergencies.

Where that money is going is visible in the line items. The FY 2027 approved budget adds $5,521,230 for additional technology costs for the Over Watch and Logistics (OWL) Unit "to enhance safety, security, and technology," and $377,168 for three security threat group regional coordinators. The same approved budget cycle directs GDC to "explore all options for additional closed-security, single-cell inmate capacity and report to OPB and chairs of Appropriations Committees" — a directive to expand restrictive housing capacity. The corresponding investments in rehabilitation are dwarfed by comparison: $39,786 for additional programming at Metro Reentry Facility in FY 2027, and in the FY 2026 amended cycle, $150,000 for a peer-led programming pilot at Autry State Prison and $93,179 in additional programming dollars at Metro Reentry. The ratio — millions for surveillance and segregation, tens of thousands for programs — is the warehouse, expressed as accounting.

What the Dorms Actually Look Like

The author who writes under the byline Trigger Cat describes Pulaski State Prison between 2023 and July 2025 in terms that match the budget pattern. In a GPS Tell My Story post titled "The Fire Alarm Kept Ringing and No One Came," she writes that "the security bubble was empty. There were no officers stationed in the dorms. We went for hours with no supervision. When something happened — a medical emergency, a fight, someone overdosing on K2 — other inmates had to call their families and have them call the facility to send help. That's how we got help. We called our mothers." She describes fights that lasted more than thirty minutes, with "blood, urine, and other fluids left on the floor," followed by mass punishment of entire dorms — loss of commissary, lockdown — that fell hardest on people who had not been fighting. Block movement, the structured time when incarcerated people are released from dorms to attend medical, dental, mental health, and education appointments, was canceled so routinely that the facility eventually had to assign a single officer to circulate dorm-to-dorm collecting people for missed mental health appointments. She names specific dates — May 15, 2024; May 16; May 19 — when no officer came for first or second block, and medical, dental, and education were missed across the board.

The author who writes as Mikemike, 32 years into a sentence that began when he was 17, describes the same absence as a survival problem rather than an inconvenience. In "Magazines Wrapped Around My Chest," he writes of wearing boots into the shower for traction, sleeping with a knife at his side, and sleeping with magazines wrapped around his chest to prevent stabbing. He recounts using an illegal cell phone to call administration to report that another older incarcerated man was dying — and waiting 41 minutes for staff to arrive at the door. "He died three minutes before they entered," Mikemike writes. "After seeing that happen, I knew there was nowhere you're safe, that if I was to die, it would be alone with no help, no officers, no medical, no nothing." The same phone, he writes, became how he saved others — keeping pressure on wounds, helping carry people to medical carts, doing the work the absent staff did not. The author Bandit, in "We Are People, Not Statistics," describes intake at Georgia Diagnostic and Classification State Prison (GDCP) in Jackson: a CERT member who threw his medical file in the trash, ignored a deputy's documented safety threat, and ordered him to strip with more than 100 other men on a 35-degree morning. He was placed in a cell where, he writes, there was "fresh blood everywhere."

These accounts share a structural feature: when staff are absent, governance does not disappear. It privatizes. The people inside become responsible for each other's medical emergencies, security, and intake processing. The agency's $1.78 billion budget purchases the building and the perimeter; it does not purchase functional supervision inside the buildings.

Education Withheld as a Tool of Control

The agency's published policies describe robust educational and rehabilitative programming. GDC SOP 108.04, "High School Equivalency Testing Centers," effective May 1, 2025, establishes procedures for administering GED testing inside facilities. SOP 503.01, "Faith and Character-Based Initiatives," effective August 7, 2025, establishes 12- and 24-month character development programs in designated dormitories. SOP 106.11, "Religious Accommodations," effective May 21, 2025, governs worship services and dietary accommodations. On paper, the rehabilitative architecture exists.

The testimony describes a different distribution. Mikemike writes plainly: "I'm a lifer so they don't like to give us education. They'll put short timers ahead of us on the list for education. I guess they feel that a person who is eligible for parole within the next three or four years is more important than a person with a life sentence that there's no guarantee is getting out. They don't try to rehabilitate you. I honestly believe they don't ever intend on letting me out so there's no reason for me to have an education other than what they need me for — sweeping and mopping floors or pushing food trays. It seems easier to control a dumb person and keep them in check." The author who writes as KingdomMan32 describes earning a college degree in Christian ministries as one of 30 students selected out of 50,000 incarcerated people — a vanishingly small allocation in a system whose own snapshots confirm that population figure. The author Wynter, sentenced in 2008 to 25 years without parole, finished his entire case plan within two years and has worked law library, education, and vocational jobs, graduated two faith-and-character programs — and writes that "nothing helps to reduce my time… The violent people are rewarded, while people like me who try to be good are punished and killed." Bandit's narrative records what intake at GDCP did to his medical file: it threw it away.

Education and programming, in these accounts, are not denied because they do not exist. They are denied because eligibility is gated by sentence length, behavior on paper that bears no relationship to actual conduct, and a rationing logic that systematically disadvantages the people who would most benefit from the chance to demonstrate change.

The Parole Board as a Second Sentencing Court

A substantial share of the Tell My Story corpus concerns the Georgia State Board of Pardons and Paroles, and the accounts are remarkably consistent in their description of a body that has replaced parole review with re-sentencing. The author CAGED first appeared before the Board in 1992 after serving the seven-year statutory minimum on a life sentence. The first denial cited "compatible with the welfare of society." The second cited "the nature and circumstances of my offense." After thirteen years inside, he was denied for eight more years — a hit so severe his counselor was afraid to deliver it, and his mother called the warden assuming her son must have done something terrible. The warden told her he hadn't, and that "the Board was supposedly doing a lot of lifers the same way."

The author GeorgiaLifer, more than 40 years into a seven-year-tariff life sentence for a single count of murder, describes a similar pattern over a longer arc. At sentencing, he writes, the average first parole for malice murder was "a smidgen over 11 years" and 83 people a year were paroled at the seven-year mark. He has been set off "like 15-16 times" since his initial eligibility, with one-year set-offs for the last eight years. He learned through informal channels that retroactive guidelines from the victim's services office were being applied to his case because the victim's family — including a prominent attorney — was protesting his release. The Board never disclosed that directly. The author Naive 00, 26 years in, has had one disciplinary report — for a cell phone, ten years ago. His first two parole reviews produced no interview at all, just denials. His subsequent interviews were not with Board members but with parole investigators, and each denial returned the same boilerplate: "Insufficient amount of time served to date given the nature and circumstances of your offenses." The author Livingwaters records that exact phrasing — "After a review of the TOTALITY of your case, insufficient amount of time served to date, given the nature and circumstances of your offense" — appearing word-for-word on his denials going back to 2017, with one-year set-offs every year. He also documents that the prosecutor in his case signed a sworn statement in February 2000 acknowledging that no evidence of force was presented at trial, and that Brewer v. State (1999) and Luke v. Battle (2002) make that evidentiary gap legally fatal — facts the Board, he writes, has refused to weigh. The author Amismafreedom describes playing his part because his family still has hope, "even though I know parole in Georgia is a joke."

These accounts converge on a structural diagnosis: the Board does not exercise the discretion the legislature delegated to it. It re-litigates the offense, applies retroactive guidelines, and uses boilerplate language that prevents any meaningful preparation for the next hearing. As CAGED puts it, "they were acting like another court instead of a Parole Board." The legal architecture of indeterminate sentencing — the seven-year tariff, the 14-year guideline, the 30-year minimum — assumes a Board capable of recognizing change. The testimony GPS has collected describes a Board that recognizes nothing but the original indictment.

Mandatory Minimums and the Death of Incentive

Wynter states the policy logic plainly: "Mandatory minimum sentencing with no possibility of parole is cruel and unusual. It takes away the one thing that might make a person want to change — hope. And without hope, the system gets exactly what it seems to want: the worst version of who we could be." His account of completing his case plan, working multiple jobs, graduating programs, and watching all of it amount to nothing in terms of release pairs with Forever19's account of receiving seventeen years at age 19 for an armed robbery that netted $140 split with a co-defendant — seventy dollars that cost him his twenties and thirties. Forever19 describes a year of sexual exploitation by an older incarcerated man at Smith State Prison, a violation he never reported because "in prison, you deal with stuff on your own. You don't ever want to be labeled a snitch, even if something happens to you personally." That, too, is a product of the warehouse: in a facility without functional supervision, reporting victimization is itself dangerous, and the survival logic of the dorm replaces any rehabilitative framework the SOPs claim to offer.

The author Amismafreedom draws a sharp contrast between two eras of Georgia incarceration. At Lee Arrendale State Prison (Alto) in the early 1990s, all movement required officer escort, and officers were physically violent in enforcing authority. At Ware State Prison after 1997, prisoners moved freely, officers were "more professional and personable," and incarcerated people kept their own property and commissary. He is not nostalgic for either — he is describing variation in how the system can be run. The warehouse is not inevitable. It is the result of choices about staffing, programming, classification, and the cultural posture of the institution toward the people inside it.

What "End the Warehouse" Would Require

The transformation plan the testimony implies is not radical in concept. It is the inverse of the budget choices the legislature has approved. It would reverse the FY 2027 directive expanding closed-security single-cell capacity. It would redirect the $5.5 million OWL technology line and the security threat group coordinator funding into the kind of programming represented by Metro Reentry's $39,786 line — but at orders of magnitude greater scale, and not concentrated in a handful of designated facilities. It would dismantle the eligibility gates that route education preferentially to short-timers and exclude the lifers who, the testimony shows, are precisely the population most invested in rehabilitative opportunity. It would replace the Board's boilerplate denials with disclosure-grade explanations — what was considered, what would change the outcome, when the next review will occur and on what criteria. It would acknowledge that Brewer and Luke mean something for cases like Livingwaters', and that retroactive application of victim-services guidelines, applied in secret, is not parole review.

Above all, an end to the warehouse would mean reckoning with the gap between the agency's policy library and its operational reality. GDC SOP 209.09 establishes the Tier III Special Management Unit as a "13-month minimum incentive-based program" with five phases of progressively fewer restrictions. The published architecture of the program reads as rehabilitative. The testimony from inside the facilities where it operates does not. Until the dorms have officers in them, the bubbles are staffed, the medical files survive intake, the education slots reach the people serving the longest sentences, and the Board hears actual change — the policy library is decoration. The warehouse will continue, and roughly 50,000 Georgians will continue to live inside it.

Sources

This analysis draws on thirteen firsthand narratives published in GPS's Tell My Story project at gps.press/tellmystory, written by incarcerated authors including Dena Ingram, KingdomMan32, Mikemike, Forever19, Bandit, Livingwaters, GeorgiaLifer, Trigger Cat, Naive 00, Wynter, Anon 30097, Amismafreedom, and CAGED. Budget figures are drawn from the Governor's Budget Report, Amended FY 2026 and FY 2027 (p. 152), and HB 974 (FY 2027G) — Senate Appropriations Committee Substitute (p. 75). Population figures are drawn from GDC's weekly Friday statistical snapshots, canonically published at gdc.georgia.gov/research/statistical-reports. Policy citations are to GDC Standard Operating Procedures 204.09, 503.01, 106.11, 108.04, 209.09, and 205.13, accessible via the agency's PowerDMS portal.