Artificial intelligence (AI) is no longer a futuristic concept. For investigators today, it is a critical tool for managing the growing flood of digital evidence. From phones and laptops to cloud platforms and social media, agencies are handling terabytes and sometimes petabytes of data in a single case. Without AI, sorting through that information would take months or even years.

In Cellebrite’s AI Advantage: Managing Investigations webinar, Lee Perlis, Director of Cellebrite’s Investigation Unit product portfolio sat down with Stephen Taylor, Vice President and General Manager of Cellebrite’s Investigative Unit, and Lt. Jarod Abshire of the Calcasieu Parish Sheriff’s Office to understand how AI is helping investigators manage digital evidence more efficiently and responsibly.

Below are highlights from their conversation.

Q: AI can sound abstract. What does it actually look like in a real investigation?

Lt. Jarod Abshire: From the law enforcement side, AI is essential now. We are not talking about replacing investigators. We are talking about helping them handle the amount of data we are dealing with. In one case, we had more than 20 terabytes of data. AI helped us quickly isolate what mattered most while minimizing exposure to harmful material such as child sexual abuse material (CSAM). It is about speed, safety and accuracy all at once.

Stephen Taylor: Pathfinder, our investigative analytics solution, has been using AI since 2016. We have seen it change what is possible for investigators. There are cases where reviewing all the evidence manually would have taken six to eight weeks. With Pathfinder’s AI capabilities, agencies were able to get through that same data in just a couple of days. One agency had already solved its case but ran the data through Pathfinder later. The system surfaced a chat message they had not seen before, one that identified the suspect almost immediately.

That is what AI can do. It accelerates the process, connects the dots and ensures investigators do not miss key details hidden in huge datasets.

Q: How does AI help reduce that exposure to disturbing content?

Lt. Jarod Abshire: In our unit, we handle cases that involve very distressing material. The toll that can take on investigators is real. With tools like Inseyets, AI can classify or flag that kind of media automatically. It helps us review what is necessary without exposing our people to every image or video. It is not about skipping steps. It is about using technology responsibly to protect your team while maintaining accuracy and integrity.

Stephen Taylor: That is right. We built these features to keep people in control but reduce risk. Assisted review ensures investigators can validate every AI-generated insight and trace it back to the source data. Nothing replaces human oversight, but AI helps filter and prioritize so the human effort is focused where it matters most.

Q: There is always concern about whether AI analysis is admissible in court. How do you make sure it stands up?

Stephen Taylor: Admissibility comes down to three things: transparency, explainability, reproducibility and human oversight. Cellebrite’s AI features are assistive, not autonomous. Every step is logged, and investigators can see exactly where each insight comes from.

When defense attorneys ask how an insight was generated, you can show them the process. That is what makes it defensible. It is transparent and fully auditable.

Lt. Jarod Abshire: That is key. We never just take AI at its word. Everything that it surfaces, we verify. That validation is what makes the results credible and court-ready.

Q: How do you prevent investigators from over relying on AI?

Lt. Jarod Abshire: It starts with training and culture. We remind our team that AI is an assistant, not an investigator. It helps us get to the truth faster, but it does not decide cases.

You have to validate, double-check and always apply professional judgment. We tell our analysts not to skip steps, to use AI to see patterns they might have missed, but to always confirm with their own eyes.

Stephen Taylor: Exactly. We encourage workflows where investigators review the full dataset, not just what AI highlights. Think of it as a force multiplier. It speeds up discovery but never replaces the human element.

Q: What is next for AI in digital investigations?

Stephen Taylor: The next step is integration. We are connecting multiple data sources, including mobile devices, computers, cloud data and CCTV, so AI can create complete and defensible timelines.

The more data investigators have, the more important it becomes to have technology that connects those dots responsibly. We are committed to keeping investigators in control. Every insight remains explainable, traceable and under human supervision.

Lt. Jarod Abshire: I think we are heading toward smarter, safer investigations. AI is not the future anymore. It is the present. It is helping us deliver justice faster while protecting our people.

Final Thoughts

AI is becoming a trusted partner in digital investigations. It accelerates review, enhances accuracy and protects investigators from unnecessary exposure to harmful content. But as both panelists emphasized, it is most effective when used responsibly, with transparency, validation and human judgment guiding every step.

As Lt. Abshire summed up, “AI helps us find leads faster, protect our teams and strengthen our cases, all while keeping people in control. That is the future of responsible policing.”

Learn More

Explore how Cellebrite’s AI-powered solutions empower investigators to manage evidence responsibly and effectively.
👉 The AI Advantage Series

Share this post