(Image credit: MrWashingt0n/Pixabay.)
Insurance companies are racing to implement artificial intelligence (AI) to process claims faster, improve consistency, and cut costs. Executives see competitors adopting AI and feel pressure to keep pace. But there’s a question most companies aren’t asking: what happens to the adjusters and claims professionals who actually do the work? What are the broader implications of AI adoption for human employees and their psychological well-being? This is particularly relevant now as burnout and mental health concerns are prevalent throughout the workforce.
Studies consistently show that 40-50 percent of employees feel burned out at work. The American Psychological Association found that about 3 in 5 employees report negative impacts of work-related stress, including lack of interest, motivation, or energy. Mental health-related absences have increased significantly, with some estimates suggesting they account for 12 billion lost working days globally each year.
When thinking about AI implementation, companies generally take one of two approaches.
The first approach aims to maximize automation, minimize human involvement, and optimize for measurable efficiency. This achieves short-term cost reduction but risks undermining the human capabilities that make insurance work effective.
The second approach uses AI to handle routine work while augmenting human capabilities in areas requiring judgment, creativity, and authentic connection. This takes more careful thought and metrics beyond efficiency, but produces organizations where skilled people want to work and can do their best work.
To understand why this distinction matters, we need to consider what makes work psychologically motivating and sustainable. Psychologists have identified three fundamental needs for psychological well-being and optimal functioning in life, including the workplace:
- autonomy(feeling in control of your choices)
- competence(feeling capable and effective)
- relatedness(feeling connected to others).
If any of these needs are not supported or are undermined, then motivation and performance suffer. AI can both support and undermine these needs, and the way human-AI work systems are designed can have a significant impact on how these new technologies affect businesses.
This matters because work is a core human activity, not just a means for producing goods and services. It shapes identity and provides purpose and meaning in life. People develop skills through work. They solve problems, learn from experience, and get better at things that matter. Work requires making decisions, applying judgment, and shaping outcomes. It builds relationships, fosters collaboration, and creates community around common goals. These aren’t just nice-to-haves; they’re core human needs. When work fails to meet these needs, people suffer psychologically, regardless of whether they’re meeting productivity benchmarks.
So what happens to these fundamental needs when AI enters the workflow? Let’s look at each one.
Autonomy
Autonomy gets undermined when AI systems are opaque and adjusters cannot understand or question them, yet must follow AI recommendations that affect their performance metrics. The adjuster feels coerced into following the AI’s recommendation even when their experience suggests otherwise. However, autonomy can be expanded when AI handles routine tasks such as data entry, document processing, and scheduling, giving adjusters time to focus on complex cases needing judgment and empathy.
Competence
When skills atrophy without use, competence can begin to erode. If AI handles initial claim evaluation, adjusters lose practice making those assessments. If it drafts communications, they lose practice crafting them. Adjusters become operators of systems they don’t understand rather than skilled professionals exercising judgment.
Yet AI supports competence when it provides comprehensive information instantly. AI that analyzes communication patterns and provides feedback on empathy and active listening could develop adjuster skills and boost professional development.
Relatedness
This suffers when AI adoption increases productivity expectations (more claims per hour), leaving less time for rapport-building. When AI scripts communications, claimants sense they are not getting authentic attention. Within teams, automation can create isolation as collaborative problem-solving diminishes. But relatedness benefits when AI handles administrative tasks, allowing adjusters more time for building relationships with claimants, having meaningful conversations, and providing emotional support.
The challenge, then, is designing AI systems that enhance rather than erode these psychological foundations.
This starts by asking the right questions up front. What aspects of claims work do employees find meaningful? What provides professional satisfaction, develops genuine competence, and requires judgment and creativity?
Putting This into Practice
Design AI to augment these elements rather than optimizing purely for efficiency. Make systems transparent and overridable. Users need to understand AI reasoning and have the discretion to disagree when their judgment differs.
Protect opportunities for skill development. Even as AI handles routine work, employees need ongoing practice with complex situations that maintain their professional competence.
Measure what actually affects outcomes like employee satisfaction, quality of relationships with claimants, development of expertise, and feelings of autonomy. Deliberately preserve relational aspects of work.
The Road Ahead
Insurance companies face a choice about what kind of organizations they’re building. Using AI for routine work while augmenting human capabilities is more strategically sound than simply maximizing automation. The reason is simple: insurance outcomes rely on human factors that AI cannot replicate.
Companies that recognize this will create better workplaces and achieve improved outcomes. Those that don’t may optimize their operations into dysfunction, one efficiency metric at a time.