This guest article from Friendly first appeared HERE
Like many claims professionals, I had developed a trusted process over the years—structured, thorough, and experience-driven. I wasn’t expecting that a new technology would make me reconsider some of that. But AI did just that.
Now comes the real expertise: the assessment requiring critical thinking.
You read all documents—often multiple times through the same set, each time extracting different details for your claim summary and decision. You're running through mental checklists:
Is eligibility met?
Any non-disclosure?
Event definition satisfied?
Applicable exclusions?
This requires holding various pieces of information in working memory to compare, contrast, and connect relevant bits for deciding claim validity. It's challenging to work in a distracting environment—phones, messages, emails, and follow-ups. Easy to lose focus when complex work demands sustained attention.
Claims professionals excel at handling chaos by creating mental structures and checklists. We develop systematic processes that become almost ritualistic.
I remember my early days as a new assessor, overwhelmed by terminology and processes. I made a crucial mistake on a disability claim involving a hereditary condition. Unfamiliar with non-disclosure concepts, I failed to properly assess the chronological medical history and missed critical family history that should have been disclosed. A senior assessor mentored me through this mistake, providing a simple checklist: "Five things you need to check on every claim." That mental structure stays with me today, regardless of claim complexity.
We're trained to identify information gaps and inconsistencies. As the saying goes, "a lie isn't only what you say—it's also what you don't say." In claims, this is particularly relevant to non-disclosure assessments. We seek evidence to validate statements, flagging gaps for investigation. Once there is sufficient evidence to support the facts, decisions must be made—whether with initial requirements or after additional information requests.
Think about assessing a disability claim involving a degenerative disc disease that leads to psychological issues and other co-morbidities. Assessors must understand these medical connections and validate how various conditions relate to claimed disabilities. When discrepancies appear, we immediately investigate:
What are the actual medical reasons for claimed limitations?
Is there information that might be missing?
We dig to find underlying causes. This requires reading all presented information while identifying links, connections, and discrepancies. We analyze medical reports against insured claims, employer information, occupation requirements, and policy restrictions to create a coherent, plausible narrative.
Transitioning to an insurance technology company, I was tasked with training colleagues on claim assessment. I created a comprehensive presentation breaking down our sequential process from administrative checks, policy verification, claim summarization, policy definition review, and assessment against policy terms.
Then the company's founder simply said, "Our AI does that all at once."
This shift—from sequential to simultaneous—fundamentally changed my perspective. I had assumed AI basically did everything humans did, just faster. But AI doesn't need sequential steps in the same way—it can retrieve and cross-reference information rapidly across multiple data points simultaneously. What struck me was AI's ability to make instantaneous data connections rather than requiring sequential comparison of information.
This revelation forced me to evaluate exactly which information relates to decision-making and how those connections work. While humans often review documents multiple times, relying on memory to connect details, AI can take a more systematic approach to data organization. You can define specific data points needed, and the AI can extract and organize this information from unstructured documents, such as lengthy reports or handwritten forms.
In other words, AI can support information processing in ways that complement human assessment—with notable speed and systematic consistency.
Simultaneous processing offers potential advantages that could address some of the limitations of traditional sequential processing. First, speed—potentially faster than the traditional, step-by-step approach. Second, consistency—more systematic in certain data extraction and organization tasks. In some AI systems, uncertainty isn't a dead end—it's an opportunity for collaboration. When confidence is low, the system can flag these areas for human review. It can highlight discrepancies and contradictions that might otherwise require multiple document reviews, potentially supporting more thorough data processing. Of course, AI systems require proper oversight and validation, and regulatory compliance remains paramount in all claims decisions.
Claims expertise remains invaluable. While AI can support by surfacing patterns, highlighting anomalies, and accelerating routine data tasks, it's the experience, intuition, and contextual understanding of claims professionals that ensure decisions are fair, accurate, and empathetic. AI works best when it supports human experts—amplifying their capabilities with better data organization, not replacing their judgment. In complex, sensitive, and often deeply personal claims scenarios, it's the trained eye and human decision-making that make all the difference. And yet, sequential thinking certainly isn't obsolete—there are many instances where traditional approaches remain essential, particularly for novel or complex cases that require nuanced judgment.
My vision centers on AI handling routine data organization efficiently, freeing assessors to focus on what they do best—examining files with their expertise and engaging empathetically with claimants. The goal is combining AI's data processing capabilities with human expertise for interpreting, empathizing, and deciding. It's about supporting people—both the professionals making decisions and the individuals behind every claim—with tools that could enhance clarity and workflow efficiency.
Among claims professionals, I often see two groups: those who feel uncertain about AI and those who are curious about its potential. My advice is simple—lean into the curiosity. AI is already entering our industry, and understanding it now helps you stay informed.
Start observing tools that might already be available to you. As you work, ask yourself: "Could AI help with the data organization aspects of this task?"
Begin to identify which elements involve routine data processing versus those requiring human judgment, nuance, and empathy.
Consider your workflow with this mindset and you might be surprised by what's possible.
For claims professionals ready to explore this evolution, the opportunity is significant. AI supporting data processing, combined with human expertise for interpretation and empathy, could create a partnership for delivering efficient yet deeply human claims experiences.
The question isn't whether AI will impact claims processing—it's how we as professionals can help shape that evolution thoughtfully.