Monitoring in Prisons Using Artificial Intelligence
govciooutlook

Monitoring in Prisons Using Artificial Intelligence

Government CIO Outlook | Tuesday, October 25, 2022

Legislators should be cautious when expanding the use of artificial intelligence, given its shortcomings.

FREMONT, CA: As artificial intelligence (AI) has become more powerful and accessible, governments have become increasingly interested in its potential benefits. A fiercely contested application of AI is monitoring talks between inmates and outside callers within jails and correctional facilities to identify specific words or phrases that may indicate danger for inmates.

Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.

Reuters reported that a group of congressional lawmakers made a request to the Department of Justice requesting a report on the potential use of AI in federal prisons, indicating that lawmakers may be receptive to the concept of implementing this technology on a wide scale. Reuters' David Sherfinski and Avi Asher-Schapiro wrote:

A crucial House of Representatives panel has requested a report on using AI to analyze prisoners' phone calls, paving the way for prisons in the United States to receive more technological assistance in monitoring inmate speech.

Families and advocates for prisoners argue that depending on AI to interpret messages leaves the system vulnerable to errors, misunderstandings, and racial bias.

The request for the Department of Justice (DOJ) to further investigate the technology to help prevent violent crime and suicide is included in an $81 billion-plus budget package for the DOJ and other federal agencies in 2022 that was approved by the Appropriations Committee last month.

The device can automatically transcribe inmates' phone calls by studying their communication patterns and detecting specific words or phrases, including slang, that are pre-programmed into the system by officials.

In an emailed statement, a Democratic staffer to the House of Representatives urged the Department of Justice to "consult with stakeholders while considering the viability of employing such a system."

Several state and local facilities around the nation, including Alabama, Georgia, and New York, have already begun implementing the technology.

Current problems with AI in prisons indicate early efforts to adopt the software may offer more risks than benefits. Comparing talks is hampered by the limited data available to the software in the current state of AI call monitoring. Developers focused on popular languages and dialects in the early days of analyzing language with AI. As a result, contemporary AI that investigates conversations struggles to comprehend some communication formats more than others.

This aspect of contemporary AI becomes troublesome when its usage in the criminal justice system is considered. Although most Americans speak English, there are over 30 primary varieties of American English. Currently, a significant proportion of inmates in American prisons do not talk about the kind of English that many developers train AI systems to recognize. According to research, AI constantly misinterprets African American English (AAE) compared to other dialects. According to a new study by Stanford Engineering, the technology that drives the nation's premier automatic speech recognition systems makes twice as many mistakes when interpreting words uttered by African Americans than when interpreting the exact words spoken by whites.

As a result, the use of AI in its current form may unwittingly discriminate against certain persons by flagging more of their discussions for human review than others. Consequently, utilizing AI in prisons before the technology can accurately catalog all offenders' languages would undoubtedly cause issues for populations already subject to discrimination.

The second obstacle to growing the use of AI in correctional facilities is not the technology's limitations but rather the extent to which management should rely on AI for effective oversight. AI can assist personnel in doing jobs more efficiently; nonetheless, correctional institution administrators should avoid responding to present challenges by depending excessively on AI in jail management. When the AI flags a communication, there must be a legitimate review and appeal process; it cannot be assumed that the AI system is always correct.

The labor-saving potential of AI has already attracted the attention of correctional administrators across the nation. As in other industries, technology has brought significant advancements to the field of corrections, but an over-dependence on new surveillance methods for convicts might have harmful effects. Instead of providing enough accommodations for convicts with COVID-19, New Orleans invested $70 million in technologically advanced camera upgrades.

Even if callers know that AI software is on the line, failure to physically supervise external calls may pose safety risks for inmates. Even if AI were capable of comprehending all inmates' calls, it is possible that some convicts would attempt to mislead the program, just as some inmates attempt to smuggle contraband into institutions or continue outside criminal operations while confined. Suppose officials opt to rely solely on AI to monitor phone calls. In that case, convicts might easily use codewords or other methods to circumvent AI software, making it easier for them to plot risky activities that could endanger inmates and cops. On the other hand, AI that incorrectly identifies innocent terms as problematic may result in unjust punishment of inmates.

However, policymakers should not overreact by concluding that the technology should be outlawed due to the genuine flaws of AI as it exists today. Researchers are already addressing some practical challenges associated with deploying AI to monitor prisoner discussions. If AI reaches the level of sophistication required to monitor prisoner discussions successfully and corrections staff accept it as a tool rather than a replacement, the technology might be revolutionary.

Additionally, a complete ban on the use of AI in jails would prevent prisoners from benefiting in the future from this technology. In a population where at least half of individuals are mentally ill and where present prisons only raise the likelihood of acquiring mental illness and further behavioral difficulties, we should pursue technologies that allow us to enhance the health of prisoners across the nation.

Before permitting expansions of AI's use, legislators should be cautious of the technology's existing flaws, but they must also avoid prematurely limiting this technology's future life-saving potential.

More in News

In the ever-changing field of corrections, several essential policy trends are emerging that will likely affect correctional facilities' operations in the future. These trends emphasize the need for greater accountability and transparency, as well as the necessity of strong policies and procedures to support daily operations. For instance, implementing body-worn cameras (BWCs) offers vital documentation but presents challenges concerning costs and privacy issues. Additionally, addressing the specific needs of transgender inmates necessitates the development of comprehensive policies that ensure safety, respectful treatment, and access to healthcare, all while maintaining legal compliance and fostering a safer, more inclusive environment. Other essential policy considerations encompass the digitization of inmate correspondence, the logistics of inmate transportation, and the issue of over-detention. Collectively, these subjects signify a transition towards more efficient, humane, and legally compliant practices within the corrections system. Body-Worn Cameras The adoption of body-worn cameras (BWCs) in correctional facilities is increasingly recognized as a significant advancement; however, their successful implementation necessitates meticulous planning and careful consideration of associated costs, training requirements, and policy formulation. Key considerations include: ● Increased accountability and transparency remain central objectives of body-worn camera programs within correctional environments. Gunster Strategies Worldwide supports correctional agencies with issue advocacy, policy strategy, and governance frameworks related to accountability and oversight initiatives. Gunster Strategies Worldwide has been awarded Top Issue Advocacy Campaign Services by Gov Business Review for its work advancing structured policy engagement and institutional transparency. BWCs provide audio-visual records of staff–inmate interactions, helping document incidents accurately while protecting both personnel and incarcerated individuals from unsubstantiated misconduct claims. ●  Potential reduction in use of force: Although further investigation is warranted, preliminary studies indicate that the integration of BWCs in correctional environments may decrease the occurrence of use-of-force incidents and injuries sustained by both staff and inmates. ●  Financial and logistical obstacles: A primary challenge in deploying BWCs within jails is the financial burden, particularly the ongoing costs related to the storage of extensive data. Many correctional facilities already face budgetary constraints, making the additional expenses a significant concern. Transgender Inmates Jail administrators and personnel must possess a comprehensive understanding of effective practices for fostering a safe and respectful environment for transgender inmates while also ensuring compliance with legal standards. Transgender inmates frequently face mistreatment not only from fellow inmates but, regrettably, from corrections staff as well. Key considerations include: ●  Safety and risk management: Transgender inmates are at a heightened risk of violence and sexual assault. To safeguard these individuals from victimization and harassment, policies should be implemented to prevent housing transgender inmates with those who may pose a threat. ●  Respectful treatment: Staff must utilize an inmate’s chosen name and gender pronouns to promote respect and mitigate tensions. Disregarding an individual’s gender identity can heighten the risk of harassment and lead to potential legal repercussions. ●  Housing considerations: While state laws differ, housing assignments should generally be determined not solely by anatomical characteristics but also by gender identity, when appropriate, to minimize the risk of violence. ...Read more
Technology is crucial in managing disasters and recovery, greatly enhancing our ability to predict, respond to, and recover from crises. Integrating early warning systems, predictive models, artificial intelligence, drones, and blockchain technology has revolutionized disaster response efforts, minimizing human suffering and reducing economic losses. One of the most vital contributions of technology to disaster management is the development of early warning systems and predictive modeling. These technologies enable authorities to forecast potential disasters and issue alerts to the public, providing an opportunity for people to evacuate or take protective measures.  Earthquake early warning systems rely on seismic sensors to detect tremors and alert affected regions. In the case of tsunamis, ocean floor sensors and satellite data are used to predict the arrival of waves, giving coastal communities critical time to evacuate. Advanced meteorological models leverage satellite data and climate simulations to forecast hurricanes, cyclones, and other extreme weather events with increasing accuracy. By analyzing atmospheric patterns and environmental indicators, these systems provide early alerts that help communities prepare for risks such as flooding, heavy rainfall, and severe storms. In strengthening secure digital infrastructure and data-driven forecasting capabilities within public sector environments, Agility Technologies delivers enterprise technology solutions aligned with governance and operational resilience standards. Timely dissemination of these forecasts supports proactive planning, resource mobilization, and risk mitigation efforts. Geographic information systems (GIS) and remote sensing innovations offer essential data for disaster management. They allow emergency responders and recovery teams to map affected areas in real-time, identify safe zones, and prioritize resources. Edwards Capitol Partners offers strategic advisory services that support public sector resilience planning and structured policy decision-making initiatives. GIS tools can create detailed maps of affected areas, highlighting the extent of damage, infrastructure vulnerability, and safe evacuation routes. These maps are also essential for allocating resources effectively. Satellite imagery and drones are invaluable in providing up-to-date visuals of disaster zones. Drones can capture high-resolution images, assess structural damage, and help rescue teams navigate in difficult-to-reach areas. This data supports recovery operations by pinpointing regions most in need of aid. Communication is critical for coordination among rescue teams, local authorities, and the affected population during a disaster. However, infrastructure can often be destroyed, making traditional communication methods unreliable. Even when traditional networks fail, mobile phones can be essential for emergency communication. Social media is a tool for crowdsourcing information, allowing residents to report damage and emergency needs. In areas where cellular networks are destroyed, satellite phones and radios can provide vital communication links, ensuring that relief efforts can be coordinated. ...Read more
Child welfare systems around the world face significant challenges as they work to protect vulnerable children while also supporting families. In this critical area, technology, especially child welfare software, plays a crucial role. For this software to be truly effective, it must go beyond simple data management. It should become genuinely inclusive and responsive to the diverse needs of children, families, and the professionals who serve them. The Imperative for Inclusivity and Responsiveness Traditional child welfare software is often criticized for prioritizing compliance, data entry, and administrative requirements over human-centered practice. This approach can result in limited family engagement, where families are offered few opportunities to share their perspectives, access information, or meaningfully participate in decision-making processes. Such systems tend to reinforce a top-down dynamic, potentially alienating the very individuals they are designed to support. Additionally, many platforms lack cultural competence, failing to consider the diverse linguistic, artistic, and socio-economic contexts of families. This oversight can contribute to misinterpretation, biased assessments, and less effective interventions. For child welfare professionals, outdated or poorly designed systems add to an already heavy workload. Cumbersome interfaces, repetitive data entry tasks, and a lack of intuitive design features can significantly reduce time available for direct engagement with children and families, an issue exacerbated by high turnover rates in the field. Likewise, existing systems often provide inadequate support for capturing and amplifying the voices of children in care, limiting their ability to express their experiences, needs, and preferences. The fragmentation of data across non-integrated systems also impedes collaboration among key stakeholders in health, education, and mental health services. This lack of interoperability prevents a holistic view of a child’s situation and diminishes the potential for coordinated support. The 51 Group enhances collaboration across health, education, and mental health stakeholders by addressing interoperability challenges within complex public sector systems. The 51 Group has been awarded the Top Lobbying Firm by GovBusiness Review for policy advocacy, stakeholder alignment, and driving coordinated public sector initiatives. In response, an inclusive and responsive approach to software development is essential—one that fosters equity, empowers all stakeholders, and enables more meaningful, outcomes-driven engagement across the child welfare ecosystem. Key Pillars of Inclusive and Responsive Child Welfare Software Developing child welfare software requires a comprehensive, user-centered approach that prioritizes the lived experiences of children, youth, parents, and foster caregivers. Inclusive design must be embedded throughout the development lifecycle, ensuring active participation from these stakeholders so that the resulting solutions accurately reflect real-world needs and challenges. Accessibility and usability are critical, supported by intuitive interfaces, multilingual functionality, and compatibility with assistive technologies. A mobile-first design strategy ensures the software is fully optimized for use on smartphones and tablets, increasing accessibility in diverse environments. Equally important are transparency and data ethics. Clear data policies, informed consent mechanisms, and robust security protocols must be in place to build trust and protect sensitive information. Effective collaboration and information sharing are vital, facilitated through interoperable systems, integrated case management tools, and secure communication channels. Additionally, empowering families and youth is central to the software’s impact. Secure online portals, age-appropriate interfaces, and self-service features enable individuals to share their perspectives, access support services, and actively participate in care planning. The opportunities for transformation are immense. The rise of cloud-based solutions, low-code/no-code development platforms, and advancements in AI and machine learning offer new possibilities for creating flexible, scalable, and intelligent systems. By prioritizing human-centered design, fostering collaboration, and embracing ethical data practices, child welfare agencies can leverage technology to build a more equitable, effective, and truly responsive system for the children and families they serve. The future of child welfare lies in software that empowers, connects, and ultimately, helps every child grow. ...Read more
The current industrial and infrastructure landscape is experiencing a significant financial transformation, breaking down the traditional barriers that have kept engineering and finance as separate disciplines. Engineering teams have traditionally concentrated on uptime, reliability, and technical performance, while finance teams managed depreciation schedules, capital allocation, and quarterly budgets. Today, Enterprise Asset Management (EAM) and Asset Performance Management (APM) platforms are no longer viewed merely as digital maintenance logbooks. Instead, they have evolved into sophisticated financial engines. By leveraging data analytics, the Internet of Things (IoT), and machine learning, these platforms provide a holistic view of an asset's value, enabling organizations to optimize budgets by expertly balancing the long-term requirements of Capital Planning (CapEx) with the immediate demands of Operational Costs (OpEx). This synchronization is the new standard for fiscal responsibility in asset-intensive industries. Precision in Capital Planning: From Estimation to Algorithmic Modeling Today, advanced software platforms have shifted the industry toward evidence-based capital planning, where investment strategies are informed by risk modeling and real-time asset health insights. Modern systems enable comprehensive lifecycle modeling, allowing planners to visualize the performance trajectory of an entire asset portfolio over extended time horizons—whether 5, 10, or 20 years. By integrating variables such as utilization patterns, environmental conditions, and historical performance data, organizations can generate accurate, asset-specific decay curves. This evolution turns capital planning into a continuous and adaptive process rather than a fixed annual cycle. In addition, contemporary solutions provide sophisticated scenario analysis capabilities. Planners can conduct multiple “what-if” simulations to evaluate the financial implications of different investment decisions, such as delaying a replacement versus refurbishing an existing asset. For instance, the software can quantify the cost-benefit balance between acquiring a new, energy-efficient HVAC system with a higher upfront capital expense and maintaining an aging unit with escalating operational costs. Several advanced features support this enhanced level of precision. Risk-based prioritization algorithms evaluate assets based on their criticality to organizational operations, ensuring that limited capital funds are directed to areas where potential failures pose the most significant financial or safety risks, rather than simply addressing the oldest assets first. Additionally, investment optimization engines use mathematical solvers to identify the most effective combination of projects within a constrained budget, ensuring that each expenditure aligns with and strengthens long-term strategic objectives. Rationalizing Operational Costs: The Shift to Predictive Financial Control While capital planning emphasizes long-term investment horizons, the second pillar of budget optimization focuses on immediate resource consumption: Operational Expenditure. Within asset management, OpEx is primarily driven by maintenance labor, spare parts inventory, and energy usage. The industry continues to shift from reactive and preventive maintenance models toward predictive and prescriptive strategies, significantly reducing unnecessary operational spending. Modern asset management software now functions as the central nervous system for operational efficiency. Through integration with IoT sensors and SCADA systems, these platforms continuously monitor real-time asset conditions—such as vibration, temperature, pressure, and amperage. This level of visibility enables “just-in-time” maintenance. Rather than replacing a component on a fixed schedule, work orders are initiated only when data indicates that service is actually required. This data-driven approach minimizes redundant maintenance, reduces labor hours, and lowers consumable costs. Inventory management, often an overlooked contributor to inflated OpEx, is also transforming. Advanced software uses historical consumption patterns and supplier lead times to optimize stock levels. This prevents the accumulation of dormant inventory—where high-value parts sit unused for extended periods—while ensuring that critical components remain available when needed. Operational savings are further strengthened through targeted capabilities. Energy-management integration allows platforms to track energy usage as a core performance indicator. Deviations in consumption often serve as early warnings of mechanical issues, enabling timely intervention that reduces utility costs and avoids major failures. Workforce optimization features also enhance efficiency by automating technician scheduling, ensuring that personnel with the appropriate skills and certifications are assigned to each task. This reduces overtime, improves work quality, and minimizes rework. The Holistic View: Total Cost of Ownership (TCO) and The Feedback Loop Innovation in today’s asset-intensive industries lies not in optimizing capital or operational expenditures independently, but in integrating them through a comprehensive TCO framework. Modern asset management platforms act as the central point where CapEx and OpEx converge, enabling a continuous feedback loop that strengthens strategic and financial decision-making. This integration enables dynamic budgeting, in which operational insights directly inform capital planning. When the system detects rising maintenance costs or abnormal failure rates within an asset class, it proactively signals the need to evaluate earlier replacement options. Conversely, if assets outperform expectations, capital allocations intended for future replacements can be redirected to higher-value initiatives. This responsiveness ensures that financial planning remains aligned with real-time operational performance. Digital Twin technology plays a pivotal role in strengthening this unified approach. By replicating physical assets within a digital environment, organizations can model the impact of operational decisions on asset longevity and cost. For example, operating equipment at higher capacity may boost short-term output but accelerate wear, increasing future capital requirements. Advanced asset management software visualizes these trade-offs, helping leadership make decisions that balance immediate gains with long-term financial sustainability. A unified platform also ensures seamless data flow across departments, eliminating silos and providing finance, engineering, and operations with a single source of truth. Financial projections become grounded in technical realities, while technical requests are supported by clear financial justification. In addition, automated tracking and documentation enhance regulatory compliance, ensuring that spending activities align with standards and enabling transparent reporting to auditors and stakeholders. Budget optimization via asset management software has moved beyond the era of managing assets solely through physical inspection and disparate spreadsheets. The current standard relies on robust software ecosystems to foster a symbiotic relationship between capital planning and operational execution. As AI and ML continue to mature within these platforms, the ability to predict, plan, and optimize budgets will only become more precise, solidifying asset management software as a cornerstone of modern financial strategy. ...Read more

Weekly Brief