Matt Polega is a cofounder and president at Mark43, a leading cloud-based public safety software company.
Ask most people what police officers do all day, and they’ll describe what they see on TV: fighting crime, making arrests and solving cases. That work is important. But it’s not the full picture.
Modern policing is largely about responding to the everyday issues that shape community life: traffic crashes, mental health calls, welfare checks, neighborhood disputes and noise complaints—the calls that don’t as often make the news.
And then there’s the part almost no one sees: the paperwork.
Every call, stop, report, arrest and follow-up creates an administrative trail. Officers spend hours writing reports, pulling information from disconnected systems, re-entering the same details and checking their work. According to the 2024 Arizona Law Enforcement Retention Survey, officers spend one-quarter or more of their time on administrative tasks, far more than time spent on search and rescue, event security or community outreach.
Every hour spent in software systems or chasing down information is an hour officers aren’t spending in the community.
The Data-Rich, Bandwidth-Strapped Reality Of Policing
Modern policing generates more information than ever: body-worn camera transcripts, video evidence, computer-aided dispatch (CAD) and records management system (RMS) data, officer notes and more. More information should mean better reports, stronger cases, greater transparency and more complete records. But for too many agencies, turning that data into a complete, compliant police report is still painfully manual, time-consuming and prone to error.
The challenge only grows as police agencies face staffing shortages, increased workloads, growing call volumes and more regulatory scrutiny. Officers are expected to document incidents with greater precision than ever, while facing constant interruption. New calls come in. Priorities shift. Officers get pulled away mid-report and hours later must try to reconstruct details from memory. Even after submission, reports frequently bounce back for corrections, forcing officers to sit at their desks instead of returning to the community.
Investigators face the same problem at a different scale. A 2025 study (download required) from the city of Albuquerque found that nearly 96% of completed complaints against officers involve some form of video, most of it from on-body recording devices (OBRD). Reviewing and summarizing video can take hours, and video is just one small part of the information that investigators must analyze.
Worse yet, investigators may spend hours reviewing a report only to find a single omission or procedural issue that can undermine an otherwise prosecutable case.
The Cost Goes Beyond Productivity
The impact is bigger than lost time. When details are missing, reports move back-and-forth between an agency, the city and the courts. Investigations slow down. Cases take longer to move through the justice system. People wait longer for answers. We’re hearing this directly in the field. One agency my company worked with shared that before moving to a modern, AI-powered RMS, in one week alone, 7 out of 10 arrest reports were sent back by courts due to errors, missing details or not following guidelines.
This administrative burden also takes a toll on recruitment, retention and morale. Many officers I’ve worked with have told me that report writing is the least enjoyable part of their job. Long hours spent finishing paperwork can cut into rest, recovery and family time.
This is where AI can offer meaningful operational change, but only if it is grounded in human judgement. AI-powered automation is an effective way to reduce administrative burdens, reclaim time, decrease frustration, improve accuracy and free people to do what they signed up for with greater confidence.
My company recently surveyed first responders and law enforcement personnel. We discovered that 77% of these workers found outdated technology to be a hindrance when doing their jobs. Ninety percent of those surveyed said automating data entry could streamline their roles, and 93% of those surveyed said they would “support their agency’s use of AI to improve operations.”
Agencies are beginning to see AI for what it can be: a practical tool for helping officers and investigators manage growing complexity with more speed, accuracy and confidence.
Five Ways Agencies Can Put AI To Work
For public safety agencies, the most valuable applications of AI are often the most practical ones:
• Use AI as a powerful validation tool to strengthen data integrity and reduce errors. AI can catch and flag errors or missing information earlier in the process to avoid ripple effects in critical reporting and investigation workflows.
• Position automated validation as a real-time training tool for officers. AI can help guide officers as they write, prompting them to capture the right details, meet reporting standards and align with policy. This should not replace supervisors. AI can act as an officer’s co-pilot, reinforcing reporting standards and helping officers identify issues earlier in the process. This can give newer officers the kind of intuitive, real-time support they increasingly expect from workplace technology.
• Program AI to pull from all critical data sources to connect the dots faster. Dispatch records, officer notes, reports and body-worn camera transcripts all contain important pieces of a story. AI can bring that information together, reduce repetitive data entry and ultimately accelerate investigations.
• Transcribe and organize body-worn camera videos and audio so people can actually use it. Video evidence is valuable, but reviewing it manually can take hours. AI can help officers and investigators to quickly search, review and verify key moments without manually combing through hours of footage. Searchable transcripts can significantly reduce review time while improving accuracy.
• Embrace responsible AI practices from the beginning. In public safety, trust matters. Agencies need AI that prioritizes transparency, human oversight, privacy, compliance and integration into existing workflows. AI should support human judgment, not replace it. The people doing the work must remain accountable for the decisions that affect communities.
When agencies embrace AI with the right mindset, the benefits can be significant: more accurate reports, fewer errors, less administrative burden and faster workflows. But the real promise of AI is helping officers spend less time behind a screen and more time where public trust is built: in neighborhoods, conversations and everyday moments that shape how communities experience public safety.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
