Public Safety Agencies Rework Intelligence Operations as AI Shifts Focus From Data Collection to Human Judgment — i2 Group - August, 13 2026

State and local public safety agencies are entering a new phase of artificial intelligence adoption in which success depends less on automating investigations than on helping intelligence professionals process growing volumes of information while preserving governance, transparency and public trust. That evolution is reshaping how agencies think about intelligence sharing, analyst training and investment priorities across law enforcement, corrections, probation and other public-sector organizations.

BizTechReports recently spoke with Roger Stokes of i2 Group for an executive vidcast examining how AI is reshaping intelligence operations across state, local and education (SLED) organizations. The discussion explored how public safety agencies are balancing the operational advantages of AI-assisted analysis with the continued need for human judgment, cross-jurisdictional collaboration and governance as intelligence operations become increasingly data-driven.

According to Stokes, rather than replacing investigators or intelligence analysts, AI is increasingly being viewed as a force multiplier that accelerates data collection, correlation and visualization while leaving critical investigative decisions to experienced professionals. The result is a growing emphasis on human oversight, data integrity and governance as agencies seek to improve operational efficiency without compromising civil liberties or confidence in public safety institutions.

That balance reflects one of the central operational challenges confronting public safety organizations. Agencies are collecting more information than ever before from an expanding network of sources, yet they are under pressure to make decisions faster while operating within complex jurisdictional, legal and organizational boundaries. AI offers new capabilities for analyzing large datasets, but technology alone cannot provide the context necessary for operational or investigative decision-making.

AI Drives a Broader Approach to Intelligence Sharing

The growing volume of investigative data is prompting public safety organizations to rethink how intelligence is collected, analyzed and shared across agencies. While law enforcement remains at the center of those efforts, agencies increasingly recognize that effective public safety depends on collaboration with corrections, probation, parole and other organizations that possess relevant operational information.

AI, Stokes explained, is helping agencies create what intelligence professionals refer to as a common operating picture by bringing together information from multiple systems and jurisdictions more quickly than traditional analytical methods. Rather than replacing existing investigative processes, AI is helping analysts identify relationships across disparate data sources while allowing managers and operators to work from a more consistent view of emerging threats.

The shift is particularly significant within state and local government, where agencies frequently operate under independent governance structures and legal authorities. Sheriffs' offices, municipal police departments, corrections agencies and other public-sector organizations often maintain separate systems and reporting relationships, making information sharing a longstanding operational challenge.

AI can help bridge those organizational boundaries without eliminating the governance controls that individual agencies require. Role-based access controls, privacy protections and established investigative procedures remain essential components of the intelligence process, even as technology enables broader collaboration.

He also noted an evolution in how intelligence professionals think about information sharing. Historically, agencies often operated under a "need to know" philosophy that limited access to sensitive information. Increasingly, however, organizations are embracing what Stokes described as a "responsibility to provide," recognizing that agencies may have an obligation to share information when doing so contributes to public safety outcomes. That shift places even greater importance on governance frameworks capable of supporting collaboration while protecting sensitive information.

Human Judgment Becomes More Important, Not Less

While public discussion often focuses on AI's ability to automate analytical work, Stokes argued that the technology is making experienced intelligence professionals more valuable rather than less.

He described AI as particularly effective at processing large volumes of information and accelerating analytical workflows. What it cannot provide, however, is the operational context required to transform information into actionable intelligence. That responsibility continues to rest with trained analysts, investigators and operational leaders who understand investigative practices, legal requirements and the broader circumstances surrounding individual cases.

Maintaining that human role also serves a broader purpose beyond operational effectiveness. According to Stokes, public trust depends on citizens understanding that AI assists intelligence professionals rather than replacing their judgment. Human oversight provides accountability while helping ensure that investigative decisions remain transparent, defensible and consistent with established governance policies.

As AI capabilities expand, the required skill set for intelligence analysts is also changing. Stokes identified three areas that are becoming increasingly important.

The first is data integrity. Analysts must understand where information originated, how it was collected and whether it can be trusted for investigative purposes.

The second is AI literacy. Intelligence professionals need to understand how AI-assisted tools generate analytical outputs, what processes those tools perform and where human validation remains necessary.

The third is governance. Analysts must become familiar with evolving frameworks governing AI use in public safety, including guidance addressing privacy, civil rights, civil liberties and responsible AI practices.

Those competencies, Stokes suggested, position analysts to validate AI-generated insights while ensuring that technology remains aligned with investigative standards and public expectations.

The same principles apply to concerns surrounding AI hallucinations and algorithmic bias. Rather than presenting AI-generated intelligence as authoritative, agencies should clearly communicate that technology accelerated portions of the analytical process while trained professionals remained responsible for evaluating evidence and making operational judgments.

Workforce Development Emerges as a Strategic Investment

The discussion also highlighted a broader shift in how public safety organizations may need to think about AI investments.

Technology acquisition alone, Stokes said, is unlikely to produce meaningful operational improvements unless agencies also invest in workforce development, analyst retention and long-term organizational sustainment. Implementing AI-assisted intelligence capabilities requires more than introducing new software; it requires integrating those capabilities into established investigative processes while preparing personnel to use them effectively.

That workforce challenge comes at a time when many state and local agencies continue experiencing turnover among intelligence professionals. As experienced analysts move into new positions or retire, organizations must ensure that institutional knowledge, governance practices and analytical expertise continue developing alongside technological capabilities.

Within that context, Stokes described i2 Group's Analyst's Notebook as an example of how established investigative tools are evolving to support AI-assisted workflows. Long recognized for helping investigators visualize relationships across disparate datasets, the platform incorporates AI capabilities that accelerate data integration and analysis while preserving human oversight throughout the investigative process.

Looking ahead, Stokes expects AI to become an increasingly common component of intelligence operations across public safety organizations. The objective, however, is not autonomous policing or automated investigative decision-making. Instead, he sees AI as an analytical capability that helps experienced intelligence professionals process information more efficiently while continuing to apply human judgment, maintain governance standards and protect civil rights and civil liberties.

###