Most P&C carriers are investing in AI. Fewer can say which decisions it’s meant to improve.

The story usually sounds like this: a claims triage pilot, an underwriting assistant in testing, a company-wide chat tool rolled out to employees, and a steering committee that meets once a quarter. Each piece makes sense on its own. Put together, they rarely add up to anything a CFO can tie to loss ratio or expense ratio improvements.

That gap matters more every quarter. Carriers that treat AI as a set of experiments will spend real money and learn some lessons. Carriers that tie AI to specific decisions, owned by specific teams, measured against specific outcomes, will pull ahead in the places that compound: pricing accuracy and speed to market.
‍

Why "a few pilots" is not a strategy

Pilots are useful. They are also easy to start and hard to turn into ongoing work. Without a clear link to business priorities, they tend to stall in one of three ways:

• No owner. The pilot sits between IT, a business unit, and an innovation team, so no one is accountable for scaling it.
• No baseline. Nobody measured how long the work took before, so nobody can prove the tool made it faster or better.
• No decision attached. The tool produces output, but no one uses it to make a pricing, product, or underwriting decision.

A strategy starts with the decisions that move results, then asks where AI can make those decisions faster or better, then funds the work accordingly.
‍

Five questions to pressure-test your plan

If you lead product, pricing, or underwriting at a carrier, these are worth answering with your team before the next budget cycle.

1. Which decisions are we trying to improve?

Efficiency gains are real, but they are only half the value. The bigger prize is better decisions: sharper rate relativities, better-targeted endorsements, fewer filings that come back with objections. List the decisions first. Then map AI use cases to them. If a use case does not connect to a decision, question why it is funded.

2. How well do you actually know what competitors are doing?

Competitive intelligence informs many pricing and product decisions. Yet most teams still assemble their competitive view by hand: pulling SERFF filings one at a time, opening rate manuals and actuarial memoranda, and rebuilding tables in spreadsheets. That process is slow, and it can leave gaps in what your team knows about competitors and when they know it. An AI strategy that ignores competitive intelligence is building on a weak foundation.

3. Where is your investment going, and why there?

Look at the split across underwriting, claims, distribution, and product. Look at which lines of business are getting attention. Then ask whether that mix reflects your strategy or simply reflects which vendors called first. A carrier growing in small commercial should not have the same AI strategy as one repositioning a personal auto book.

4. What will you build, and what will you buy?

General-purpose AI tools are good at general-purpose work. Insurance filings are not general-purpose. They are long, dense, inconsistently formatted, and full of tables that matter down to the decimal. Decide where in-house effort adds unique value (your own data, your own models) and where a purpose-built platform gets you there faster. Then make sure you have the internal skills to support that mix as it evolves.

5. Can you explain the output?

Regulators, auditors, and your own actuaries will ask how a number was reached. Any AI you use in pricing and product work needs to point back to its source: the specific filing, exhibit, or page. Transparency is not a compliance afterthought. It is what lets your experts trust the tool enough to use it.
‍

Why product and pricing is the right place to start

Many carriers begin their AI work in claims or customer service. Those are reasonable choices. But product and pricing teams offer a practical starting point: time savings you can measure immediately, in work that informs rate and product decisions.

Consider how much of a pricing analyst's week goes into finding and reading competitor filings rather than acting on them. Locating a competitor's latest rate action. Hunting for the rating algorithm inside a 300-page rate manual. Rebuilding a multi-tier factor exhibit from a scanned PDF. That work is necessary, repetitive, and a strong fit for AI. Cut the time it takes, and your experts spend more of their week on judgment: deciding what to file, where, and why.

It is also easy to measure. You know how long a competitive review takes today. You can measure how long it takes next quarter.
‍

How Insuraviews fits into your AI strategy

Insuraviews is a regulatory filing intelligence platform built for P&C product, pricing, and compliance teams. Our AI Assistant works directly on the full contents of rate and form filings, including every attachment.

In practice, that means your team can:

• Find the section that matters. Ask for a competitor's classification schedule, rating algorithm, or actuarial justification, and go straight to it.
• Get usable tables. Rate relativities, class code schedules, and multi-tier factor exhibits are extracted and reconstructed into a form you can work with, not retyped by hand.
• Trace every answer. Results point back to the source filing and exhibit, so your actuaries and compliance staff can verify what they are looking at.

The result is a competitive review that can drop from 20 minutes to 2. Across a product team, over a year, that adds up to faster rate actions, better-informed filings, and experts spending their time where it counts.

It also gives your broader AI strategy something every strategy needs: a clearly owned use case, a measurable baseline, and a direct line to pricing decisions.
‍

The bottom line

The pace of AI adoption in P&C is not slowing down. Carriers that answer the hard questions now, about decisions, data, investment, ownership, and explainability, will be the ones whose AI spend shows up in results rather than just in the IT budget.

Competitive intelligence gives product and pricing teams a practical place to start: a defined problem, a way to measure improvement, and better information for the decisions they already make.

Already a customer? Log in at app.insuraviews.com and give it a try, and as always, let us know what you'd like to see next.

Not yet a customer? Schedule a demo of Insuraviews today to see what you're missing.

‍