DON’T FIRE YOUR FUTURE

What high-tech companies’ AI

whiplash means for independent agencies

If you automate without retraining, you’re not

trimming fat, you’re cutting muscle. And your people feel it.

By Carolyn Smith, APR, CRA, TRA

We all watched it happen. High-tech companies laid off thousands of employees in the name of efficiency, cost control and automation, calling it a “strategic realignment.” Then, quietly, they reversed course and rehired tens of thousands of new employees.

What they learned the hard way is something agencies need to recognize now: They didn’t have too many people. They had too few people who knew how to work with the technology. In the process, they stripped away judgment, cultural context, and client nuance—the kind of intelligence that doesn’t live in a dashboard.

What those employees lacked wasn’t value. It was training. So those firms cut capability and then paid to rebuild it.

You’re standing on the same fault line, asking:

• What AI tools should we buy?
• What happens to our people?

If you get the first one right and the second one wrong, you won’t win.

The productivity mirage

If you ask leaders how AI is going, you’ll hear words like leverage, scale, and productivity. If you ask your account managers, you’ll hear something else:

• “It’s confusing.”
• “It’s extra work.”
• “I don’t know what I’m allowed to put into it.”
• “It’s not really saving me time.”

That gap matters.

According to Gallup’s 2026 workforce survey on AI adoption, about 12% of U.S. employees report using AI daily in their job, and roughly 26% say they use it at least a few times per week, while many others use it only occasionally or not at all. Adoption isn’t your biggest issue. Integration is.

In agencies, the pressure is real:

• Premiums are rising
• Clients are more skeptical
• Margins are thinner
• Carriers are stricter
• Talent is harder to find

So naturally, leaders are looking at:

• AI-powered policy checking
• Automated servicing workflows
• Virtual assistants
• Proposal generators
• Data extraction tools
• Client communication bots

These are smart, necessary experiments. But they don’t always work the way you hope. You’ve felt it—stuck in a loop with a virtual assistant, repeating yourself, pressing zero, finally screaming, “Connect me to a human!” That frustration is what happens when technology promises efficiency but forgets the person on the other end.

Patricia Pike, director of agency operations for Three Arbor Insurance in Birmingham, Alabama, says something every agency owner should tattoo on their strategic plan: Virtual assistants work best when processes are mature, repeatable, and stable. If your workflows still require deductive reasoning, judgment, or nuance—if they’re filled with caveats—simply outsourcing or automating the task won’t solve the problem. It will expose it. Technology doesn’t clean up sloppy thinking; it accelerates it.

In agencies, real operational intelligence lives in places technology can’t replicate:

• The account manager who knows which client hates email and which one needs reassurance
• The CSR who remembers why endorsement wording changed years ago—and what broke when it didn’t
• The back-office professional who spots certificate errors instantly because she’s seen the same mistake hundreds of times

These people aren’t overhead. They are your early-warning system. Replace them with AI without teaching them how to supervise AI, and you won’t gain leverage; you’ll lose judgment.

Heather Parsons, people and operations manager for The Insurance Market in Salisbury, Maryland, explains, “I see firsthand how much institutional knowledge lives inside long-tenured account managers and leaders. At our agency, we have several employees who have been with us for more than 30 years. Many now serve in management roles, and others continue to work directly with clients. That depth of experience is not just tenure; it is context. It is knowing why a coverage decision was made ten years ago, recognizing patterns in claims, and understanding the nuance behind client relationships.

“When agencies terminate before they retrain, they are not just cutting costs. They are cutting context. In insurance, context is a core element of risk management. AI should enhance that judgment, not eliminate it. Upskilling preserves cultural memory while increasing execution speed, and that is a far better investment than rebuilding lost knowledge later.”

If you automate without retraining, you’re not trimming fat, you’re cutting muscle. And your people feel it.

When the back office hears “AI,” they hear “replacement.” When account managers hear “automation,” they hear “commoditization.” If you don’t address that fear directly, you won’t get adoption; you’ll get resistance.

Pike emphasizes, “It’s important that leaders help their employees understand how powerful AI can make them. By adopting and advancing the use of AI within their day-to-day workflows, they [employees] are not only reducing human error and saving time, but they are also giving themselves the freedom to do what they are best at—working directly with their clients. That’s what the employees want, and that’s what the clients want as well, as it creates the relationships that create better retention.”

Your two-year window

You likely have about 24 months before this shift shows up clearly in your financials.

• AI will mature
• Carriers will embed automation
• Competitors will experiment aggressively
• Clients will expect faster turnaround
• Younger employees will assume AI is native

The hybrid strategy that agencies need is simple:

1. Buy the right tools
2. Build the right people

Yes, explore AI for:

• Policy checking
• Endorsement comparisons
• Certificate generation
• Data extraction from loss runs
• Proposal assembly
• Renewal summaries
• Marketing content drafts

Let the machine handle repeatable, rules-based work. But retrain your people to do what machines cannot:

• Interpret
• Advise
• Contextualize
• Empathize
• Lead

Insert AI at controlled points where it reduces friction, not judgment:

• Data extraction from loss runs
• Checklist generation
• First-pass proposal drafts
• Document comparisons

Let AI carry the repetitive lift. Free your team to focus on meaning, advice, and relationships.

In practice

What does this look like for various agency positions?

Account managers—train to spot hallucinations, question overconfident outputs, and verify coverage interpretations.

• Use AI to draft renewal summaries and comparisons; you provide the advisory layer that clients rely on.
• Let policy-checking assistants flag differences; you own final accuracy and carrier negotiations.

CSRs/service advisors—train to recognize misrouting, identify complaint patterns, and escalate sooner.

• Auto-generate COIs and endorsement checklists; validate lender and project-specific requirements.
• Use governed search across SharePoint or AMS; confirm applicability and flag exceptions.
• Build a human off-ramp into every client channel.

Producers/advisors—train to research and interpret trends.

• Use AI to assemble proposal shells; you lead trade-off and strategy conversations.
• Leverage AI for prospect briefs; you decide what matters.

Claims support (agency-side)— train to recognize red flags and escalation triggers.

• Use AI to structure timelines and document requests; you manage expectations and intervene when files stall.

Ops /compliance—train to demand explainability and governance.

• Stand up a lightweight AI use program: purpose, data boundaries, bias checks, vendor oversight, and monitoring.