Skip to sources
Time Machine

Correctional Officers and Jailers

Scrub through 219years of this role's history, from when it first emerged, through every wave of technology that reshaped it, to the cited projections for where it's heading next.

2026drag to travel through time
18251850187519001925195019752000now
Country
2026
Known today as Correctional Officers and Jailers (BLS SOC 33-3012)
US Employment
381K
OEWS is a point-in-time survey snapshot, not a continuous time series; BLS advises against using it for year-over-year trend comparison.
Median Annual Wage
$58,940
≈ $57,429 in 2024 dollars
Each dot is a cited figure over time; the dotted line only links them (values between aren't measured). Hollow dots are estimates.
Tools of the era

The tools that defined the work

Select an era to see how it reshaped the work.

  • Iron key + strap + lockstep (the keeper's kit)

    The Auburn system's correctional officer — called a keeper or turnkey — had three instruments of work: a large iron key for the cell locks, a leather strap for corporal punishment of silence violations, and his own physical presence enforcing the lockstep formation in which prisoners walked with their hands on the shoulders of the man in front. The architecture of Auburn and Eastern State Penitentiary (1829) was itself a tool: Auburn's radial cell-block design allowed a single officer to stand at the center and observe multiple tiers; Eastern State's panopticon-like corridors permitted visual surveillance of individual cells. These two architectural templates — Auburn's congregate-labor design and Eastern State's solitary-cell design — shaped every US prison built for the next century. The keeper was the enforcement node of an architectural system designed to minimize the number of eyes needed to watch the maximum number of prisoners.

    Effect on the work

    The silent system's enforcement structure required relatively few keepers per prisoner by later standards: Auburn operated in the 1820s-30s with a staff of roughly 30-50 overseers for populations of 700-900 prisoners. The labor of prisoners in the workshops generated revenue that partially offset staffing costs — a model explicitly designed to make punishment economically self-sustaining.

    Work toolChanging equipment
  • ACA Declaration of Principles (1870) + telephone call system (early 1900s)

    The National Prison Association's 1870 Declaration of Principles — the founding document of what became the American Correctional Association — introduced a concept that had not previously been applied to prison guards: the reformatory ideal. If the goal of incarceration was reformation of character rather than simple punishment, then the people responsible for the confined had a professional obligation, not just a custodial one. The declaration called for officers to be trained, compensated fairly, and selected for character. In practice, these ideals were largely ignored for decades — prison guard jobs were patronage appointments in most states through the 1930s — but the document established the philosophical vocabulary that professionalization advocates would use for the next century. Parallel to this, telephone intercom systems (installed in larger prisons by the 1910s-1920s) allowed central control rooms to coordinate guard positions across cell blocks without requiring physical runners.

    Effect on the work

    The reform impulse of the 1870-1940 period produced the first rudimentary training programs in some states (Massachusetts and New York were early movers) but had minimal effect on national staffing ratios or officer qualifications, most of which remained informal and politically appointed through the New Deal era.

    Work toolChanging equipment
  • Two-way radio + central control room + perimeter towers (post-WWII modernization)

    World War II-surplus radio technology entered prison management in the late 1940s. Two-way portable radios — the same technology that had transformed police patrol — gave guards in large prisons a way to communicate in real time across sprawling cell blocks and yards without physically running to a post telephone. Central control rooms, equipped with radio consoles, wall-mounted annunciator panels, and in later years closed-circuit television monitors, became the nerve centers of modern correctional facilities. Perimeter guard towers, which had been a standard feature since the 1800s, were equipped with rifle-armed officers and communication links to the control room. This era also saw the first systematic use of CCTV in correctional settings — primarily at entry points, control rooms, and sally ports — driven by the same analog camera technology being deployed in banks and industrial facilities.

    Effect on the work

    Radio communication allowed somewhat higher guard-to-prisoner ratios on large yards and corridors, as officers could summon backup rapidly. It did not reduce the absolute number of officers needed; it improved their response capability. The expansion of state prison populations through the 1950s and 1960s meant officer headcount grew steadily during this period regardless of technology changes.

    Work toolChanging equipment
  • Post-Attica professionalization — training standards, merit hiring, ACA accreditation (1974)

    The Attica Prison riot of September 9-13, 1971 — in which 43 people died (33 inmates and 10 correctional officers and staff), all but one guard and three inmates killed by law enforcement gunfire during the state police retaking — forced a reckoning with correctional officer professional standards that the 1870 Declaration of Principles had articulated but never enforced. New York Governor Rockefeller's subsequent commission, and parallel state-level investigations elsewhere, documented that officers had received minimal training, were poorly supervised, had no clear use-of-force protocols, and had no institutional channel for raising safety concerns. The ACA founded its Commission on Accreditation for Corrections in 1974, establishing the first national voluntary standards for prison staff selection, training, and supervision. By the 1980s, most state systems had moved from patronage hiring to civil-service merit systems; pre-service training requirements (typically 120-200 hours of academy instruction) were codified in most states. Correctional officer unions (AFSCME and state-specific bargaining units) expanded rapidly in this period, driven by the danger of the post-Attica environment and the organizing momentum generated by the riot's casualties.

    Effect on the work

    Post-Attica professionalization increased the per-officer cost of the workforce (higher salaries, training requirements, union contracts) without reducing the number of officers needed. The mass-incarceration wave of the 1980s-1990s ensured that any efficiency gains from better training were overwhelmed by the requirement for more bodies as prison populations grew.

    Work toolChanging equipment
  • Digital CCTV + electronic key + biometric access (surveillance infrastructure build-out)

    The 1994 Crime Bill's $9.7 billion prison construction program generated a wave of new facility construction through the late 1990s and 2000s, and the new prisons were built with digital surveillance infrastructure from the start. Digital CCTV systems (IP cameras connected to DVR servers) replaced analog camera-and-VCR installations; electronic key systems (proximity cards and electronic locks, replacing physical keys for most interior doors) were installed in new and renovated facilities; biometric hand and fingerprint readers controlled access at sally ports. Electronic control of cell doors — previously operated by a officer physically pulling a lever or turning a key — transitioned to push-button electronic actuators at the officer's post, then to computerized master-control systems. A single officer at a central console could now manage cell-door operations for an entire housing unit that previously required officers at multiple posts.

    Effect on the work

    Electronic door control and digital CCTV enabled modest reductions in posts required for routine cell management and corridor surveillance at modern facilities. In practice, the staffing gains from technology were largely absorbed by the growth in incarcerated population through this period; the net officer headcount continued rising through 2008.

    Work toolChanging equipment
  • COMPAS risk-assessment algorithm (Northpointe, deployed 2000s) + ProPublica 2016 exposé + Wisconsin v. Loomis 2016

    COMPAS (Correctional Offender Management Profiling for Alternative Sanctions), developed by Northpointe (later Equivant), was adopted by Florida's Broward County for pretrial risk assessment in 2008 and spread to dozens of state systems through the 2010s. The algorithm scored defendants on recidivism risk using 137 factors including criminal history, age, and social circumstances — and these scores were used by judges to inform bail, sentencing, and parole decisions. On May 23, 2016, ProPublica published "Machine Bias," an investigation that analyzed COMPAS scores for more than 7,000 Broward County defendants. The finding: Black defendants were labeled high-risk at nearly twice the rate of white defendants; Black defendants were 77% more likely to be flagged as future violent criminals; and the algorithm proved only 61% accurate — barely better than chance — for the violent-crime prediction that carried the most sentencing weight. In 2016, Eric Loomis challenged a Wisconsin state court's use of his COMPAS score at sentencing; the Wisconsin Supreme Court ruled in 2016 that using the algorithm did not violate due process, reasoning that judges used it as one factor among many. The US Supreme Court declined review. The COMPAS controversy established the central tension in algorithmic criminal justice: efficiency and consistency on one side; opacity, racial disparity, and the defendant's inability to challenge a proprietary algorithm on the other. For correctional officers, the practical effect was that the risk label attached to each person they supervised now arrived from a software system rather than a clinical assessment.

    Effect on the work

    Risk-assessment algorithms did not directly reduce correctional officer headcount; they reshaped how officers understood their charges. A prisoner labeled "high risk" by COMPAS received more restrictive housing and fewer program opportunities — decisions made upstream of the floor officer but affecting the daily population the officer managed.

    Work toolChanging equipment
  • First Step Act + PATTERN risk-assessment tool (Federal BOP, 2019) + body-worn cameras (~50% adoption by 2024)

    The First Step Act, signed December 21, 2018, required the Federal Bureau of Prisons to develop a new risk-and-needs assessment tool within 180 days. The DOJ announced PATTERN (Prisoner Assessment Tool Targeting Estimated Risk and Needs) in July 2019 — designed to classify all federal inmates on recidivism risk and match them with appropriate rehabilitation programs. Civil rights organizations raised immediate concerns about PATTERN's potential to perpetuate racial disparities; the DOJ responded with multiple revisions through 2020. For the first time, a federal statute required that risk-assessment scores be used not merely for security classification but to allocate rehabilitative programming — expanding the algorithm's role from predicting who might reoffend to prescribing who gets what intervention. Body-worn cameras, which had transformed policing after Ferguson (2014), began entering corrections more slowly: correctional environments present different challenges than street patrol (confined spaces, shower areas, medical care), and adoption lagged police by several years. By 2024, approximately half of state correctional systems had deployed some body-worn cameras, primarily for use-of-force documentation.

    Effect on the work

    PATTERN and body-worn cameras both added administrative work to the correctional officer's role: PATTERN-driven programming required officers to escort inmates to new rehabilitation activities and to document compliance; body cameras required uploading footage and created accountability for use-of-force incidents that had previously gone undocumented. Neither reduced the floor-officer headcount requirement.

    Work toolChanging equipment
  • COVID staffing crisis + AI behavior-detection CCTV + decarceration policy pressure (the contraction era)

    COVID-19 arrived in US prisons in March 2020 and spread rapidly through congregate facilities. The Marshall Project and Associated Press tracked 398,627 prisoner infections and 2,715 prisoner deaths by mid-2021; 114,237 correctional workers were also infected and 209 died. Staff infection rates were 2-3 times the general population rate at many facilities. The crisis accelerated early releases — the US incarcerated population fell by approximately 230,000 between March and December 2020, the largest single-year decline in US correctional history. The staffing crisis that followed was partly pandemic-driven and partly structural: officers who left during COVID found better-paying, lower-risk work elsewhere; agencies reporting 15-30% vacancy rates by 2022 could not easily hire back into a profession defined by physical danger, shift work, and chronic understaffing. The Federal Bureau of Prisons reported approximately 15% vacancy rates in 2022-2024. Simultaneously, the technology layer continued advancing: AI behavior-detection CCTV (systems that flag anomalous movement, potential fights, or falls without a human monitor watching) were adopted by some state systems in this period, though the correctional environment's particular challenges (controlled movement, multiple prisoners in frame, privacy constraints) slowed commercial deployment compared to hospitals or commercial buildings.

    Effect on the work

    The COVID era produced the first significant involuntary contraction of the correctional officer workforce since mass incarceration began: not through technology displacement but through pandemic-driven population decline, staff departures, and the inability to recruit replacements at wages competitive with the tightened post-COVID labor market. BLS projects a further -7% decline through 2034 as decarceration continues and AI tools modestly reduce monitoring labor.

    Work toolChanging equipment
Projection cone · present → 2034

What credible sources project

Scrub the slider past now to anchor each scenario on the scrubber. The spread is the range of futures credible sources project for this role.

Employment outlook
Projected change in the number of people doing this work.
Staffing-crisis floor scenario (vacancy gap stabilizes, slow attrition)
2034
-3%
Optimistic scenario: the staffing crisis of 2020-2024 is resolved through wage increases (some states enacted 15-20% correctional officer pay increases in 2022-2024 specifically to stem departures), improved officer wellness programs, and prison population stabilization as decarceration plateaus. If the vacancy rate recovers to pre-COVID levels (~5-8%) and the incarcerated population stabilizes near 1.8-1.9 million, the headcount decline would be modest — approximately -3% from 2023, driven primarily by continued AI-assisted monitoring reducing some fixed-post requirements at newer facilities. This is the bull case for employment stability: the occupation contracts slowly rather than sharply.
BLS National Employment Matrix 2024-34
2034
-7%
BLS Employment Projections 2024-34 cycle — the most current. Projects a -7% decline in correctional officer employment from the 2023 base, equivalent to approximately 27,400 fewer positions by 2034. This is the central BLS estimate; the OOH attributes the decline to continued decarceration (criminal justice reform legislation reducing sentence lengths and expanding alternatives to incarceration) and modest productivity gains from AI monitoring tools. Despite the net decline, BLS projects approximately 26,500 annual openings through replacement need — the occupation's high turnover rate means significant hiring even in a shrinking field. The -7% projection does not model speculative scenarios; it models current decarceration trends and policy trajectories.
Continued-decarceration pessimistic scenario
2034
-18%
Scenario analysis: if the decarceration trend accelerates beyond BLS baseline assumptions — through federal First Step Act implementation reducing federal populations further, state-level marijuana legalization reducing drug-offense incarceration, and continuing bail reform reducing pretrial jail populations — the incarcerated population could fall from approximately 1.9 million (2023) toward 1.4-1.5 million by 2034. At historical staffing ratios, that population decline would imply approximately 20% fewer officer positions than the 2023 baseline. The -18% estimate represents the mid-point of this scenario, assuming AI monitoring tools modestly improve staffing efficiency while decarceration removes the primary demand driver. This is the pessimistic tail of the cone.
AI task exposure
Share of the role’s tasks that researchers estimate AI can do. This is a measure of task exposure, not a forecast of jobs lost.
Frey & Osborne (2013)
2033
35%
of tasks
Gaussian-process classifier on O*NET task features. Frey & Osborne assigned Correctional Officers a probability of computerization of approximately 0.60 — placing them in the moderate-risk tier of the 702-occupation dataset, meaningfully above police patrol officers (~0.098) but well below security guards (~0.84). The bottleneck factors that held the score below extreme-high risk: higher social intelligence requirements than security guards (de-escalation in a confined environment with a captive population is demanding), significant physical manipulation (pat-downs, cell extractions, restraint), and the legal authority requirements that no robot can currently hold. The 0.60 score reflects F&O's view that routine monitoring and documentation functions could be automated while floor-officer presence could not. The -35% figure translates the F&O probability into an approximate employment ceiling if fully realized over two decades; BLS's actual -7% projection suggests F&O's implied automation timeline is optimistic.
Eloundou et al. — 'GPTs are GPTs' (2023)
2028
4%
of tasks
GPT-4 task-by-task LLM exposure labeling on O*NET tasks. Correctional officers have moderate LLM exposure on the documentation, report-writing, and incident-logging tasks that consume a significant share of officer time — an AI analog to Axon's Draft One for police could reduce this burden. Core floor-officer tasks (inmate count, pat-down searches, cell extractions, de-escalation, movement supervision) are not text-based tasks an LLM can perform. The -4% estimate represents the realistic ceiling of LLM substitution: report automation and AI-assisted behavioral monitoring, not physical floor replacement. Eloundou et al. classify correctional officers as an occupation where LLM exposure is moderate on the administrative component and near-zero on the physical-presence component.
Today, in this role

What's shifting in the work right now

The historical view above shows how this role has moved. This is the present-day detail: which AI tools are picking up which tasks, where the edge still is, and the natural directions this work can grow.

What's changing in your day

Three parts of your work where AI is already doing real lifting, and what stays yours.

AI is sitting alongside you hereMonitor AI-flagged alerts from communications surveillance platforms (phone call transcription and behavioral scoring systems) for evidence of planned violence, gang activity, contraband coordination, or criminal enterprise

Monitor AI-flagged alerts from communications surveillance platforms (phone call transcription and behavioral scoring systems) for evidence of planned violence, gang activity, contraband coordination, or criminal enterprise; route credible leads to investigators and document the chain of custody.[5],[12]

Tools picking this up
Where your edge is

Treat AI-flagged communications as intelligence, not evidence: the model flags patterns it was trained to associate with risk, but its error rate is not publicly disclosed and algorithmic bias in predictive systems is a documented concern. Document your reasoning when you act on or dismiss a flag. This creates a defensible record and protects you if the flag turns out to be wrong.

AI is sitting alongside you hereProcess intake and release paperwork, maintain accurate housing assignment records, enter program participation and disciplinary data into the jail management system, and retrieve records in response to internal and legal requests.

Process intake and release paperwork, maintain accurate housing assignment records, enter program participation and disciplinary data into the jail management system, and retrieve records in response to internal and legal requests.[10],[2]

Where your edge is

Treat the jail management system as the authoritative record. Discrepancies between the system and physical reality (a person recorded as housed in one unit but physically in another) are a safety and legal liability. Your job is catching those discrepancies, not data entry for its own sake -- the software is getting better at data entry faster than it is at physical verification.

AI is sitting alongside you hereWrite incident reports documenting use of force, rule violations, injuries, and unusual events

Write incident reports documenting use of force, rule violations, injuries, and unusual events; generate logs that comply with facility, state, and federal evidentiary standards and that will withstand scrutiny in disciplinary proceedings or civil litigation.[6],[10]

Where your edge is

Use AI-assisted report generation to pull timestamps, camera references, and sensor data automatically, but never submit a report you have not reviewed word-for-word. Incident reports are legal documents; inaccuracies expose you personally and your agency institutionally. Accuracy review is a skill that becomes more important, not less, as AI drafts the first pass.

Where this role is heading

Natural next steps for someone with your foundation: not exits, evolutions.

A direction you could grow

Probation Officers and Correctional Treatment Specialists

Correctional officers who have worked directly with incarcerated people carry direct experience in behavior assessment, documentation, de-escalation, and navigating the criminal justice system -- all core to probation work. The pivot typically requires a bachelor's degree in criminal justice, social work, or psychology, which many COs pursue while working. The transition moves officers out of a facility into community supervision, substantially reducing physical risk and mandatory overtime exposure while opening a path to case management and rehabilitation-focused work.

What you'd add
  • · Bachelor's degree in criminal justice, social work, or psychology (required by most agencies)
  • · Case management fundamentals: structured assessment tools (LSI-R, ORAS), supervision plans, and caseload management
  • · Motivational interviewing and trauma-informed communication for community supervision contexts
  • · Report writing at the level required for pre-sentence investigation reports and violation of probation petitions
What it takesSome new skills to pick up
Share this year
Drops anyone you send it to straight into 2026.
Preview card
Part of Protective Services · see all 14roles →
Different role?

See the same long-arc view for your own profession.

Browse the directory by industry, or search by title or SOC code. New roles ship every few weeks. Every profile cites every claim.

Browse all roles

The data behind this timeline

On record since1817
Latest tracked employment380,500 (US, 2025)
Latest median pay$58,940 (2025)
Outlook-7% by 2034 (BLS National Employment Matrix 2024-34)
View all 28 cited data points
YearUS employmentMedian annual paySource
195035,000n/aESTIMATE
1972100,000n/aESTIMATE
1980180,000$13,500ESTIMATE
1990282,000n/aESTIMATE
2000427,000$31,390BLS-OEWS
2003417,420$33,160BLS-OEWS
2004409,580$33,600BLS-OEWS
2005411,080$34,090BLS-OEWS
2006417,810$35,760BLS-OEWS
2007431,980$36,970BLS-OEWS
2008n/a$38,380BLS-OEWS
2009455,350$39,050BLS-OEWS
2010457,550$39,040BLS-OEWS
2011448,740$38,990BLS-OEWS
2012434,870$39,040BLS-OEWS
2013432,680$39,550BLS-OEWS
2014434,420$39,780BLS-OEWS
2015427,790$40,530BLS-OEWS
2016431,600$42,820BLS-OEWS
2017428,870$43,540BLS-OEWS
2018415,000$44,330BLS-OEWS
2019436,000$44,660BLS-OEWS
2020405,870$47,410BLS-OEWS
2021392,600$47,920BLS-OEWS
2022363,250$49,610BLS-OEWS
2023392,000$49,550BLS-OEWS
2024365,380$51,810BLS-OEWS
2025380,500$58,940BLS-OEWS
Embed this timeline on your site

Free for any site. Paste this where the timeline should appear; it stays interactive, every datapoint stays cited, and it sets no cookies on your page. How embedding works

<iframe src="https://futurehistory.earth/embed/33-3012"
  width="100%" height="430" style="border:0"
  title="Correctional Officers and Jailers, a Future History timeline"
  loading="lazy"></iframe>

See all roles in Protective Services