Opinion
The next investigator: From collecting evidence to understanding it with AI
As AI systems become more powerful, the importance of human judgment only increases.
Across today’s workforce, one thing is becoming increasingly clear: those who don’t learn how to work with AI will fall behind. This isn’t just about using new tools, it’s about collaborating with systems that generate insights, managing them and turning them into value. And this shift isn’t confined to tech. It is now reaching one of the most traditional and mission-critical domains: digital investigations.
Here, the implication is especially sharp: the next investigator will no longer be defined by their ability to collect evidence, but by their ability to understand it with AI. Are investigators in the digital age required to develop a new capability for working with AI systems?
For years, digital investigations have faced the same core challenge: not a lack of evidence, but an overload of it. As digital data has exploded - communications, location data, social networks - the time required to make sense of it has grown with it. Investigations didn’t stall because information was missing, but because turning it into insight took too long.
This is not a local phenomenon. In the U.S. and Europe, experts increasingly describe a “data-rich, insight-poor” reality - cases filled with massive volumes of data, yet struggling to produce timely conclusions. In Israel, where resources are often more constrained relative to the scale of data, the gap is even more pronounced.
And this is where the question becomes sharper: what happens when the challenge is no longer finding evidence, but understanding it?
The answer is a fundamental shift in the investigator’s role. They are no longer just collecting information but managing the thinking process that interprets it. And this is exactly where AI comes in.
In the traditional model, the investigator formulates a question - who communicated with whom, where the suspect was, and passes it on to analysts. They connect dots, build timelines, and identify patterns. This process takes time - days, sometimes weeks. It is the primary bottleneck in digital-era investigations.
An investigator who does not know how to work directly with AI systems remains dependent on this model. By contrast, a new generation of AI systems is changing the rules of the game. These are not tools that simply return an answer, they’re systems that help manage an investigative process. Agentic AI-based systems break down complex questions into sub-tasks - location analysis, communication cross-referencing, anomaly detection and deploy multiple “agents” in parallel, each examining the data from a different perspective.
Related articles:
The impact is already visible in practice. Investigators can ask complex questions in natural language - for example, whether a suspect’s alibi aligns with location and communication data, and receive evidence-based answers within minutes, including the identification of inconsistencies. Processes that once took weeks are now completed in a fraction of the time. It is not only an improvement in efficiency; it is a shift in the role.
Investigators are no longer waiting for analysis, they are managing it directly.
This is where the shift becomes clear. It is no longer enough to understand evidence; investigators must know how to work with systems that can help understand it. This is not merely a technical skill, it is a new professional capability - to ask the right questions, identify gaps in the answers, challenge conclusions and decide when not to trust the system. It marks a shift from execution-based work to judgment-based work.
This transformation is already reflected in the tools emerging across industry. Companies like Cellebrite, operating at the core of digital evidence, are moving from systems that present data to systems that allow investigators to interact with data directly, analyzing it conversationally and applying AI-powered reasoning to the underlying evidence itself.
But here too, it’s important to clarify: AI does not replace the investigator - it redefines the role.
Accountability stays human. AI-generated insights cannot stand on their own, especially in environments where findings may be tested in court. Investigators are still responsible for validating, explaining and ultimately making decisions.
If anything, as AI systems become more powerful, the importance of human judgment only increases.
The implications going forward are clear. Law enforcement agencies that do not train investigators to work alongside AI systems will struggle to keep up with the growing volume of data. Those who embrace the shift, by contrast, will not only shorten investigations, they will also be able to tackle cases that were previously unsolvable.
It is no longer sufficient to understand evidence; the ability to operate systems that can interpret it has become essential. It is no longer sufficient to ask questions; what matters is the ability to frame them effectively. In an era where the evidence already exists, the advantage no longer lies with those who uncover it, but with investigators who know how to work with AI.
The author is Ronnen Armon, Chief Products and Technologies Officer at Cellebrite.
