Embedded vs. bolt-on: How insurers will pursue measurable results from AI in 2026

The noise around artificial intelligence can be deafening. AI is here, but for all the noise, the fact is our industry has only begun to get a sense of what it can actually deliver. The last few years have seen many carriers and MGAs adopt AI at the edges through bolt-on tools to a legacy platform in the hopes of meaningful progress. That approach has had its day, and as a value proposition, seems to have missed the mark. Heading into 2026, we must realize that AI's greatest gains will come from transforming the core processes that define the insurance value chain and enable insurers to build the future they and their customers seek.

True opportunity is in the core, not at the edges

Consider the foundation of what we do: underwriting, quoting, claims, compliance. These are all areas where AI can do the heavy lifting for humans. Hurdles to that dream of a fully streamlined process come in the form of the legacy platforms on which these processes sit. Meaningful AI integration of these decades-old behemoths can be quite the tall order, and a bolt-on approach is likely not going to result in real value. For us to wring real value out of the tech, AI needs to be embedded directly into workflows to manage and run the process, rather than simply added on.

 But embedding AI into a 10-to-15-year-old legacy system is no easy undertaking. Carriers and MGAs need to consider what or who is controlling their systems and processes and whether they allow for adaptability. Those equipped with modern, agile digital platforms will have greater control over their processes and have the flexibility to more nimbly integrate AI functionality efficiently into their ecosystem to see how it might improve their core processes.

Enterprising carriers and MGAs with ambition looking to embrace the future will want to consider the following best practices in their approach to AI adoption in 2026:

·       Test-and-learn to foster real change
The idea that you need a massive budget to apply AI effectively is a complete misconception. Instead, think small. Embrace your ambition, but scale your initiatives modestly, so you can afford to fail and try again with a different approach.

Adopting this test-and-learn mindset reduces risk, generates feedback quickly, and encourages teams to stay nimble. It also helps carriers to avoid the expensive trap of chasing shiny objects – the tools or applications that may appear impressive on the surface but in reality, really fail to solve the problems that weigh on the business.

·       Protect the data; protect the foundation
It's commonplace that AI is only as good as the data that feeds it. In our industry, that amounts to a material risk.

 As companies consider AI, they should focus on three fundamentals:

1. Provenance matters.
2. Preserving raw data.
3. Keeping humans in the loop.

 Insurers who take this approach will be well-positioned to scale AI safely and confidently.
·       Look for the learners: they will fill the roles that don't exist yet
The feared scenario is that AI will automate entire insurance roles out of existence, but real life is much more complicated. Properly applied, AI will become the ideal assistant to a skilled professional, augmenting their efforts, while the ultimate authority remains in human hands. Accordingly, AI will give rise to jobs that have yet to fully evolve. One likely future example: a senior AI underwriter responsible for overseeing a team of AI agents.

And to find occupants for these positions that have yet to fully evolve, look around your organization for those who like to learn rather than those who like to be taught. Identify the exceptional people in your workforce who thrive on curiosity; they are gamechangers. These hardy souls will be essential to bridge the gap between human expertise and AI-driven processes.

·       Focus efforts on high-impact workflows for sustained ROI

Prioritize workflows where speed and precision are critical, such as broker submissions, automated underwriting, subrogation identification and managing renewal risks. Leverage AI agents to streamline data retrieval, decision-making and actions across systems, while baking in checkpoints for human oversight. This targeted approach ensures AI investments address key pain points, driving adoption and delivering sustained returns.

In 2026, AI will move from the margins and into the fabric of insurance, redefining how we do business. Carriers and MGAs that want to reap the benefits – streamlining operations for transformative efficiency and optimizing a more frictionless and engaging customer experience – must take practical steps now with core processes in mind.