Insurers rate themselves AI leaders but none has rebuilt distribution around it - KPMG

What happened

A new KPMG analysis finds that insurers largely see themselves as ahead of the curve on artificial intelligence, yet the industry has not restructured its distribution model to reflect that self-assessment. According to the findings, self-perceived AI leadership among insurers has not translated into any carrier fundamentally reshaping how policies are sold and serviced through agents, brokers and direct channels.

The report points to a gap between confidence and execution: insurers are investing in AI tools and describing themselves as advanced adopters, but the underlying distribution architecture — the mix of intermediaries, digital channels and customer touchpoints through which insurance is actually bought — remains largely unchanged.

Why it matters

For an industry built on distribution — where the channel through which a policy is sold shapes pricing, underwriting speed and the customer relationship itself — this is a structural, not cosmetic, gap. Bolting AI onto existing workflows (faster claims triage, chatbot-assisted quoting, internal automation) can improve efficiency without changing the experience a customer actually has when buying or renewing cover. KPMG's framing suggests that many insurers are using AI to optimise what already exists rather than to rethink how risk is distributed and sold in the first place.

That distinction matters for transformation leaders across financial services: AI maturity claims are increasingly common, but the credible test is whether a business has changed its operating model — commission structures, broker roles, digital-first acquisition — as a result. Insurance is a useful bellwether because distribution is so central to the sector's economics, making any lag between AI investment and structural change highly visible.

The Renascence take

Self-reported AI leadership is a low bar; it costs nothing to claim and rarely gets audited against outcomes. The more revealing question is always what an organisation has been willing to dismantle, not what it has added.

Most "AI leadership" in financial services today is additive, not architectural — new capability layered onto an unchanged operating model. The tell is distribution: if the channels, incentives and roles through which a product reaches a customer look the same as they did five years ago, AI is being used to defend the status quo, not to redesign the experience. A genuinely AI-native insurer would be willing to renegotiate commission structures, redefine what a broker or agent is for, and let direct and intermediated channels compete on service rather than habit. Until that happens, "AI leader" is a marketing claim, not an operating reality — and customers will keep experiencing the old model with a faster interface bolted on top.