Earning confidence in insurance AI: Ten speakers I’m watching this month

This month close to 4000 people from across the global insurance community will check into the Fontainebleau for Guidewire Connections. On behalf of GFT Technologies Canada, I’ll be one of them. 

I’m looking forward to the trip, in part because I expect that the conversations I’ll have in Las Vegas will line up closely with the ones I’ve been thinking about in Québec City and Toronto since January.

Those talks started this year with a number: while 62% of insurance executives in Grant Thornton’s 2026 AI Impact Survey described their AI maturity as scaling across multiple functions, only 24% were confident they could pass an independent AI governance review within 90 days.

That review is no longer hypothetical in either the U.S. or Canada. In the U.S., 25 states and D.C. have adopted the NAIC’s model bulletin on insurers’ use of AI. In Canada, OSFI’s revised Guideline E-23 brings AI squarely into model risk management for every federally regulated insurer from May of 2027 onward. 

Insurers operating across the border face two regimes that ask the same thing: show us who is accountable when a model gets it wrong. This, all at a time when AI is moving faster than the paperwork. 

It’s quite a time to be working in this space, to say the least.

Guidewire has put “Insurance with Confidence” on this year’s banner. The ten speakers below are the ones I think can tell us whether that confidence has been earned, and what it took.

Two Canadian carriers on the main stage

Two of the five headline speakers come from Canadian insurers, and I don’t think that’s an accident.

For example, Canada’s carriers have been quietly ahead on catastrophe modeling and climate disclosure, partly because their regulator pushed them there sooner than most. If Guidewire wanted people who’ve already lived through the problems the rest of the industry is about to face, that’s where they’d look.

Natalie Simms, VP of Operations and Modernization at TD Insurance, will speak for the country’s largest direct-to-consumer home and auto group. TD started with ClaimCenter in 2008 and began moving its InsuranceSuite to Guidewire Cloud in 2019. 

That’s nearly two decades of modernization with no broker standing between the insurer and the customer. I want to hear how she decides which decisions AI can take on when the policyholder’s only point of contact is TD itself. 

Shane Mannion, interim CIO at Gore Mutual, represents a carrier that has been early more than once. Gore was the first Canadian insurer to launch policy, billing and claims on Guidewire Cloud at the same time, in 2021, and it completed its move to the Guidewire Cloud Platform this past week. Fifteen months past that cutover is about when the honest lessons surface, and an interim CIO usually has a sharp view of what matters next.

Randy Termeer, President of USAA’s Property and Casualty Group, rounds out the insurer voices on the main stage. USAA’s members move often and expect the same service wherever they land. If AI-assisted decisions hold up at that scale and that level of member scrutiny, it tells the rest of us something. 

Two speakers who will talk about people

Author David Epstein‘s Range has circulated through our leadership team more than once. His new book, Inside the Box, published in May, argues that constraints improve creativity and judgment. Insurers live inside constraints every day, from OSFI and provincial regulators to reinsurance treaties. 

With more than half of insurance leaders naming regulation as a barrier to scaling AI, his talk may turn out to be one of the most practical of the week. A soft market is a constraint too, and I’d like to hear how he thinks teams should work inside one.

Colin Jost, writer and co-anchor of Saturday Night Live’s Weekend Update, will give the room a break from adjusted operating margins. 

I’m curious to see how much material he finds in our industry. Probably more than we’d like.

Where the delivery lessons live

The main stage sets the tone. The breakout rooms are where I usually learn the most. 

On Wednesday, October 28th Kevin Stabe, Head of IT for Midcorp at Arch Insurance, will walk through how Arch built a new technology platform during a business separation, integrating more than 60 systems and migrating thousands of policies and claims in 16 months. He’ll share the session with Michael Mahoney, Guidewire’s Chief Delivery Officer, and Deloitte’s Abhi Gupta. 

On Tuesday Sai Giridharan, CIO of Heritage Companies, will join his claims director and Hi Marley to talk about AI-enabled claims from first notice of loss to first action. After a Calgary hailstorm, the first 48 hours of intake shape everything that follows. I want to know what Heritage lets the machine decide on day one and what still waits for an adjuster.

Drew Whitmore, Global Head of Partnerships at Shift Technology, will demo cross-carrier fraud detection inside ClaimCenter on Wednesday. Organized fraud doesn’t stay within one insurer’s book, and it tends to move fastest in the weeks after a catastrophe, when every claims team is stretched.

Tony Alonzo, Genesys’s insurance industry lead for North America, will speak at the Partner Showcase Theater on the member claim experience. It’s no question that AI that makes that interaction worse will cost more than it saves.

What I’ll bring home

With Swiss Re expecting P&C premium growth to stay near a cyclical low through 2027, insurers can no longer count on rate to carry their top line. That changes the modernization question.

At the same time, this also raises the bar for AI. As more claims and pricing decisions run through models on a modern core, insurers carriers will need to show regulators, boards and policyholders not just that the AI is in use, but that it is working as intended, and why.

The carriers that answer both questions well will be the ones that grow through the soft stretch rather than wait it out. 

At the Fontainebleau, we’ll see who’s setting the pace.

Article by André Gagné, CEO at GFT Technologies Canada