Efficiency isn’t the only goal when embracing artificial intelligence. Gallagher’s Damion Walker argues that real value from AI means developing new insights, improving customer experience, and accelerating time to market.

As an industry, insurance isn’t at the vanguard of artificial intelligence. But, it just might wind up to be a bellwether

But when it comes to answering the questions every type of stakeholder is asking — How will AI transform our business? What does AI mean for my career? — the insurance industry provides the measure. 

Even as the industry embraces agentic AI and the implications of machine learning on predictive analytics, it’s unlikely an insurer is going to make headlines on unlocking a unique breakthrough in the use of large language models. 

After spending decades in this industry and watching technology trends come and go, I’ve learned that the most important issue isn’t about transformation. It’s about the application. 

AI is transforming our processes, our data treatment, and our systems. AI’s promise is one of greater efficiency for rote processes like compliance documentation, rate comparisons, renewal processing, and similar tasks. But insurance professionals are still best suited to do the human-centric work, from sales and service to analysis and claims advocacy.

I’d like to start by dispelling the big fear (and for some leaders, the biggest hope) associated with these advances. AI isn’t going to fully replace anybody in commercial insurance anytime soon, but it will repurpose time spent on repetitive administrative tasks that AI can handle more efficiently. Generally speaking, we are so understaffed across the industry, our focus should be on how AI can make our existing workforce faster, more streamlined, and driving value-added inputs for our stakeholders. 

That’s a real opportunity.

The Reality Check We Need

I hear the same thing almost every day. 

“My staff is twice as big as it used to be, but we’re handling three times as many clients.” The math doesn’t work unless we get smarter about how we operate. 

Where do we apply our staff, how do we organize our teams, when should we enhance communication between departments, and, finally, after we’ve addressed those questions, we ask how we can use AI to help speed things along?

The talent gap is particularly acute and calls for this sort of strategizing. We have a knowledge divide between seasoned professionals who understand the nuances of risk and younger staff who bring energy and tech-savviness but lack industry depth. Meanwhile, we’re operating as multinational organizations with service teams spanning continents, each facing its own turnover issues.

This is where AI becomes less about futuristic visions and more about immediate, practical solutions. How do we take the institutional knowledge that sits in the heads of experienced brokers and make it accessible to a client service manager in Mumbai or a recent college graduate in Chicago? 

That’s the dream AI application that keeps me up at night— in a good way.

The Infrastructure Challenge: fragmentation

Before we can turn AI’s fantastic hype into practical hope, we need to address a fundamental reality. Valuable industry data is woefully fragmented. 

As an example, Gallagher has acquired over 500 companies in the past 30 years. Each brought their own systems, their own data structures, their own ways of doing business. Getting all of that dispersed knowledge into one coherent, AI-accessible format isn’t just a technical obstacle; it’s an archaeological dig through decades of business evolution.

This is why AI in insurance, though rapidly maturing, is still in its infancy stage. We’re not ready for the sci-fi scenarios where a client says, “I’d like an insurance policy,” and AI handles everything from risk assessment to policy issuance. We’re years away from that. 

Honestly, I’m not sure we want to get there entirely.

What we can do right now is focus on the building blocks. How can we make policy reviews easier? How can we streamline the renewal process? How can we help our people identify the right coverage options faster? These aren’t the sexy applications everyone is dreaming of, but they’re the foundation on which everything else will be built.

The Human Element Remains Central

There’s something that gets lost in all the AI excitement: Commercial insurance will always require human judgment. Unlike personal lines, where a customer can buy a State Farm policy online with no human interaction, commercial insurance involves complex risk factors that systems don’t yet understand. Every business is different. Every risk profile has nuances. Every client relationship has a history and context that matters.

This doesn’t mean technology can’t help. It absolutely can and should. But the role of AI should be to augment human expertise, not replace it. I want technology that brings my decades of experience to my team’s desktop. I want systems that help a client service manager in India understand the risk factors that matter to a manufacturing client in Ohio. I want tools that help our people provide better products and more effective service to our clients.

The Partnership Approach

One thing that’s become clear through my conversations with startups and established vendors is that the most successful AI implementations come from true partnerships. The vendors that understand our business model — whether that’s the broker model, the carrier model, or the agency model — are the ones that can actually move the needle.

It’s not enough to have a cool AI application. Focusing on those kinds of things generates ephemeral enthusiasm that quickly dissolves into hype. We can avoid that trap by asking smarter, more practical questions. Can you tie your efficiency gains to our core KPIs? Can you show me how your solution impacts the  value of our service delivery to our customers or reduces our cost of operations? Can you walk me through the story of how pulling one lever in your system ultimately improves our entire organization and what other applications can the same deployment model improve?

The vendors that can answer those questions and demonstrate they understand insurance beyond just the technology layer are the partners that create real value.

Looking Forward Pragmatically

We all operate under the business mantra, “We need to do more with less.” And in insurance, we also need to do it better. Management is pushing for financial KPIs, cost reduction, and operational efficiency. AI can help with all of these goals, if we approach it strategically.

For startups looking to work with organizations like ours, here’s my advice: start small and prove value quickly. 

If you’re not holding confidential data or company secrets, you can get in the door more easily. Show us how your solution makes our people’s jobs easier, without making them more complicated. Understand that we’re dealing with compliance requirements, data security concerns, and integration challenges that might not be obvious from the outside.

For my colleagues in the industry, my advice is similar. Be patient with the technology, but aggressive about the application. 

We know AI won’t solve all our problems. But it can solve specific, immediate problems if we’re thoughtful about implementation.

The future of AI in insurance isn’t just about complementing professionals’ talents; it’s also about scaling it. It’s about taking what we know works and making it available to more people, in more places, at those moments when it matters. That’s a future worth building toward, one practical application at a time.