The algorithm now sits quietly inside the claims file, shaping decisions that once belonged entirely to human judgment. While it never takes the stand, its presence will be felt in every deposition, motion, and trial that follows. Adjusters now face a new reality in which they must defend not only what they decided, but also how they reached that decision with the assistance of a system that processes data faster and more broadly than any individual could. This shift creates a new kind of witness, one that does not speak but must still be explained in clear and direct terms.
Plaintiffs will not limit their focus to the adjuster's conduct; they will turn to the process itself and ask how the system works, what data it uses, and whether its output deserves trust in a legal setting. They will frame the algorithm as a hidden force that may carry bias, errors, or flawed assumptions, and argue that any decision influenced by such a tool deserves scrutiny. The adjuster must be ready to meet that challenge with clarity, because confusion will only strengthen the argument that the process lacks reliability.
This evolution changes the nature of claims litigation in a fundamental way, because the case no longer rests only on facts and judgment, but on whether the system used to evaluate those facts holds up under examination. The adjuster must now stand at the center of both the human and technological aspects of the claim, and that dual role requires preparation, awareness, and a clear understanding of how each piece fits together. The file must reflect that understanding from the beginning, because once the case reaches litigation, it will be too late to fill in the gaps.
The first challenge in these cases will center on transparency, as plaintiffs will argue that decisions influenced by AI lack the clarity needed for proper evaluation and cannot be tested meaningfully. They will describe the system as a black box that produces results without explanation, and they will suggest that such a process undermines fairness and accountability. That argument will gain traction if the adjuster cannot explain, in simple terms, how the tool contributed to the decision.
The adjuster must ground the discussion in practical use rather than technical complexity, because the goal is not to teach the mechanics of the system but to show how it fits into the decision-making process. The file should reflect that the tool provided information that the adjuster reviewed, considered, and either accepted or rejected based on the facts of the claim. This approach shifts the focus away from the system itself and back to the reasonableness of the adjuster’s conduct.
Clarity will carry the day in these cases because the side that explains the process clearly will hold an advantage. If the adjuster can walk through the tool's role step by step and show how it supported a thoughtful decision, the argument loses force. If the explanation becomes tangled or uncertain, the risk increases, because uncertainty invites doubt and doubt invites liability.
Discovery Shifts
Discovery will expand beyond traditional documents and move into the systems that influence decision-making, meaning requests will target not only emails and notes but also the tools, inputs, and outputs that shaped the claim. Plaintiffs will seek to understand how the algorithm operates, what data it relies on, and whether that data reflects a fair and accurate picture of the claim. They will attempt to pull the system into the case as a central issue rather than a background tool.
Carriers must prepare for that shift by striking a balance between protecting sensitive information and providing enough detail to explain the process, because resistance without explanation will only deepen suspicion. The adjuster cannot rely on the idea that the system speaks for itself, because it does not, and any attempt to shield it completely will likely backfire. The better approach is to focus on how the tool was used in the specific claim and why that use was reasonable.
A well-developed file will serve as the first and most effective line of defense because it can answer many of these questions before they gain momentum. If the record shows consistent and thoughtful use of AI, the need for deeper inquiry may diminish. If the record lacks detail or clarity, discovery will expand to fill that void.
The Corporate Representative
The role of the corporate representative now carries greater weight because that witness must explain not only the facts of the claim, but also the role of the systems used to handle it. This requires a level of understanding that goes beyond familiarity with the file and extends into how the carrier integrates technology into its process. The witness must be able to describe that integration in a way that makes sense to a judge or jury.
Preparation becomes critical in this context because a witness who cannot explain the system will struggle to maintain credibility. In contrast, a witness who overstates the system’s role will create unnecessary risk. The goal is to present a balanced view that shows the tool as a helpful component of a broader process driven by human judgment. That balance must come through clearly in testimony.
The representative must also be ready to address questions about consistency and training, because those issues will often form the basis of the plaintiff’s argument. If the carrier can show that it uses AI in a structured and consistent way, the defense becomes stronger. If the use appears uneven or unclear, the argument becomes more difficult to follow.
Expert Battles
Experts will play a larger role as these cases develop, because both sides will rely on them to explain or challenge the use of AI in claims handling. Plaintiffs will present experts who question the reliability of the tools and highlight potential flaws in their design or application. They will attempt to show that the system produces results that cannot be trusted.
Defense experts must respond by focusing on how the tool operates in practice and why its use in the specific claim was reasonable, rather than engaging in abstract debates about technology. They must connect the system to real-world outcomes and demonstrate that it adds value without replacing judgment. This approach keeps the discussion grounded and relevant.
The goal is not to defend the concept of AI in general, but to defend its use in the context of the claim at issue, because that is where the case will be decided. A clear, focused explanation will carry more weight than a broad, technical one.
Human Judgment
Human judgment remains the core of any defensible claim decision, and the adjuster must ensure that the file reflects active and thoughtful engagement at every step of the process. The system may provide information, but it does not make the decision, and that distinction must be clear. The adjuster must show that they considered the output, evaluated it, and reached a conclusion based on the facts.
This principle must be consistently applied throughout the file, as it underpins the defense. Notes should reflect analysis rather than repetition, and decisions should be explained in terms that show reasoning and care. The file should read as a record of deliberate action rather than passive acceptance.
Courts will accept the use of tools, but they will not accept the absence of judgment, and that line will define many of these cases moving forward. The adjuster must remain at the center of the process at all times.
The algorithm now plays a role in claims handling that cannot be ignored, and its presence will shape how cases are evaluated and defended in the years ahead. The adjuster must be prepared to explain how that tool fits into the decision-making process and why its use supports a reasonable outcome. The file must tell that story with clarity and consistency from start to finish.
The adjuster who understands this shift and adapts to it will be in a strong position to defend their work, while the adjuster who fails to do so will face increasing challenges. The difference will appear in the record and in the testimony that follows. The algorithm may not take the stand, but its role will always be part of the case.
About the Author:
Frank Ramos is a partner at Goldberg Segalla LLP. framos@goldbergsegalla.com