AI is entering justice settings quickly, yet the presence of new technology does not automatically improve practice. Tools that generate summaries or organize information may increase efficiency, but without alignment to real-world workflows, they can fall short of improving supervision, case planning, or follow-through.
At Carey Group, this shift is approached through a practical framework grounded in evidence-based practices, structured case planning, behavior-change tools, staff development, coaching, and continuous quality improvement. Carey Group supports AI adoption that starts with purpose, data awareness, human accountability, accuracy checks, population impact, and a clear test of whether AI strengthens or replaces professional judgment.
This perspective keeps AI grounded in operational reality rather than in emerging technology trends. It also highlights two areas where AI can provide meaningful support: improving how staff prepare for their work with people on supervision and strengthening how agencies develop staff skills through supervision, coaching, and review.
The benefit comes from integration with existing workflows rather than from the tool itself.
Effective AI use begins with a clearly defined operational challenge rather than with the tool itself. Justice agencies often encounter barriers such as difficulty organizing case information across large caseloads, delays in documentation, or missed opportunities to follow up on important details during supervision.
When these challenges are identified with precision, AI can be applied in ways that directly support staff performance by improving organization, preparation, and consistency. Without that clarity, however, AI often becomes another layer of activity that varies across staff and units, making it difficult to evaluate its impact on practice.
Carey Group urges the importance of identifying the purpose first by asking what problem is being solved and how the tool will improve the work staff are already expected to perform. This approach ensures that AI case management tools are selected based on their ability to strengthen existing workflows rather than operate independently from them.
Justice work depends on professional interpretation, which requires staff to consider context, behavior, motivation, and engagement rather than relying on surface-level summaries. AI can assist by organizing or synthesizing information, but it cannot take responsibility for how that information is understood or applied.
Maintaining accountability requires agencies to define who reviews AI output, how accuracy is verified, and where final responsibility remains. These expectations are particularly important when AI is used in documentation, case planning, or supervision decisions, where incomplete or misleading information can affect outcomes.
The National Institute of Justice has described AI as a tool that may support administrative and decision-related workflows within the justice system, but that support must remain bounded by professional oversight. When staff remain actively engaged in reviewing and interpreting AI output, the technology functions as a support mechanism rather than a substitute for judgment.
AI should be used responsibly in justice settings by supporting preparation, organization, and consistency while keeping interpretation, decision-making, and accountability with trained professionals. Its role is to strengthen how work is carried out, not to replace the judgment, engagement, or oversight required in effective justice practice.
Case planning is a collaborative process that depends on engagement, shared understanding, and language that reflects the individual’s perspective and readiness for change. The effectiveness of a plan is shaped by the interaction between the practitioner and the person on supervision, not simply by how complete or structured the document appears.
AI can support this process most effectively when it helps staff prepare for that interaction. It may assist in organizing relevant information, surfacing details that should be addressed, and supporting reflection on whether key areas have been considered before or after a conversation. This type of preparation strengthens the practitioner’s ability to enter discussions with clarity and purpose.
The boundary remains critical. When AI is used to generate a case plan independently, the collaborative process that builds professional alliance is removed. Plans developed without that interaction often lack the connection necessary to support meaningful behavior change, even when they appear technically sound.
With that boundary in place, AI in justice settings can be integrated into daily workflow in ways that strengthen how information moves into action. Case management depends on translating assessment results, observations, and case notes into clear next steps, and this often requires significant effort to organize and interpret information across a caseload.
AI case management tools can support this effort by helping staff consolidate information before meetings, identify areas that require attention, and assist supervisors in reviewing whether case plans and follow-up actions are aligned with identified needs. In this role, AI supports preparation, organization, and oversight while leaving decision-making and engagement firmly with the practitioner.
Carey Group’s work focuses on helping professionals apply evidence-based practices through structured tools, training, and coaching. AI can complement this work by improving clarity and consistency within existing workflows, strengthening the connection between information and action without altering the practice itself.
Introducing AI into justice settings requires safeguards that are practical enough to guide daily use. Data considerations must be addressed first, including what information can be entered into an AI system, how it is stored, who has access, and whether it can be retained or reused.
Accuracy is equally important. AI-generated content should be reviewed before it is incorporated into documentation, communication, or planning. Staff must understand when verification is required and how to compare AI output with source information.
Frameworks such as the NIST AI Risk Management Framework emphasize the importance of managing risk and ensuring that AI systems are used in a way that is reliable and appropriate for their context. For responsible AI use in criminal justice, this translates into clear expectations, training, and ongoing oversight.
Implementation determines whether these safeguards are applied consistently. Staff need guidance on how to use AI appropriately, supervisors need standards for review, and agencies need a process for monitoring whether AI is improving practice or introducing new risks. This aligns with the broader focus on continuous quality improvement that underpins effective implementation.
AI in justice settings is most effective when it enhances the consistent application of evidence-based practices without interfering with the human elements that make those practices effective. It can support preparation, improve organization, and help reduce gaps in follow-through, but it cannot replace the interaction required to build trust, encourage engagement, and support change.
Professional alliance remains central because meaningful progress depends on collaboration, understanding, and shared commitment. A case plan that reflects the individual’s voice and priorities is far more likely to support change than one that is generated without that connection.
For agencies wondering what responsible AI use in criminal justice looks like, the path forward is practical and deliberate. Define the operational need, establish safeguards, maintain human accountability, and ensure that AI supports rather than replaces the work. When those conditions are in place, AI can contribute to stronger, more consistent practice. When they are not, the risks outweigh the benefits.
Carey Group's evidence-based training and consulting services address the needs of the justice system and behavioral health professionals. Training is an essential tool for keeping staff, supervisors, leadership, and stakeholders up to date with emerging knowledge and expectations for improved outcomes. Working closely with Carey Group professionals, agencies are better able to offer a mixture of in-person, online, and self-directed courses on evidence-based practices, motivational interviewing, core professional competencies, case planning and management, continuous quality improvement, coaching, and the use of behavior-change tools and supervisor resources. Talk to a Carey Group consultant today to get started!