Will AI replace insurance brokers in NZ? What AI can and can't do for your cover

AI tools vs an insurance adviser: what each actually does

What AI is genuinely useful for when you're researching cover

AI tools are useful for the reading, not the deciding. They are best at text-processing work. Therefore, one must treat the output as a starting point, because the information may not be correct.

Three uses hold up:

• Explaining the words: Excess, co-payment, stand-down period, loading: a chatbot can put these into everyday language before you talk to anyone about health insurance, life cover, trauma insurance, or income protection cover. Policywise's income protection waiting period explained guide is an example of what that plain-English version looks like.
• Summarising a document you already hold: Your own policy wording or benefit schedule is a fair thing to paste in for a plain-English summary, though the IFSO Scheme's Ombudsman warns AI can miss key exclusions, so check anything that matters against the wording itself. That wording can carry personal and confidential details, so treat it the same way you'd treat any other personal document before pasting it in.
• Getting your questions in order: As reported by RNZ, the Banking Ombudsman says these tools can help people organise information and understand complex issues. One technique is to ask the tool to put the questions a good financial planner would ask you first.

The limit is that a large language model's major strength is predicting the next most likely word. New Zealand guidance written for people representing themselves warns that chatbot answers may be inaccurate, incomplete, or out of date, and may rest on overseas rules that do not apply here.

Where AI advice tools fall short of a licensed insurance broker

AI tools fall short at the points where money and cover change hands. A chatbot cannot judge how an insurer will underwrite your health history, argue a declined claim on your behalf, or carry responsibility if its answer turns out to be wrong. It also cannot weigh your own circumstances.

Underwriting judgement: pre-existing conditions, loadings, and exclusions

Underwriting is where the same health history produces different answers from different insurers, and a chatbot has no way to know which answer you will get. A Consumer NZ investigation using fictional applicants with declared mental health histories found materially different outcomes across 14 New Zealand providers.

Some providers would have covered the applicant's pre-existing mental health condition, while others referred it to underwriting review. Health cover stand-downs for the same profile were quoted at three years by one provider and five by another.

Across life cover, insurers also differ in how they answer perceived risk: at least one applies a premium loading rather than an exclusion. What counts as pre-existing varies too, and can turn on a symptom present when you take the policy out, rather than a diagnosis.

Having pre-existing conditions doesn't mean you can't get cover at all, or that they'll be automatically excluded from your policy. A pre-existing condition is usually excluded unless the insurer has accepted it in writing and charged an extra premium.

Policywise covers this for health insurance and life insurance separately. Your adviser can tell you how an insurer underwrites a particular condition, and whether it may mean waiting before cover starts or paying an additional premium.

Claims advocacy: who argues a declined claim for you

If a claim is declined, an AI tool can help you organise the information, but it cannot argue the case for you. A person can. If the complaint is not resolved with the provider within two months, or the two of you reach deadlock, it can go to your provider's dispute resolution scheme. For many insurers, that is the IFSO Scheme, which is free for consumers.

An adviser can pass your concern directly to the dispute resolution scheme, and some can also help at claim time. Two recent cases show what that involvement can look like:

• An income protection claim was avoided for non-disclosure. An investigation found the consumer had told the bank adviser, acting as the insurer's representative, about his headaches and an MRI before the policy was arranged. The complaint was upheld and the cover reinstated.
• A life and trauma policy was avoided over two non-disclosures. After three underwriters advised on materiality, the insurer paid half the life benefit, $95,664, to the widow.

In the year to 30 June 2025, the IFSO Scheme received 4,293 approaches and accepted 600 disputes for investigation, a 25% increase on the previous year. Health, life, or disability insurance accounted for 29% of those disputes.

This rising dispute count, with health, life and disability claims making up nearly a third of it, points to a steady stream of cases complex enough to need human judgement about materiality and underwriting practice. This is exactly the kind of judgement an AI tool cannot provide.

Accountability: who you can hold responsible if the advice is wrong

If an adviser's advice is wrong, there is a licensed person and a legal duty behind it. Anyone who gives regulated financial advice to retail clients in New Zealand must hold, or operate under, a Financial Advice Provider licence issued by the Financial Markets Authority.

An AI tool has neither.

Three statutory duties sit on the person advising you:

• Give priority to your interests, taking all reasonable steps to ensure the advice is not materially influenced by their own.
• Exercise the care, diligence, and skill a prudent person in that occupation would.
• Comply with the Code of Professional Conduct: treat clients fairly, act with integrity, give suitable advice, and make sure you understand it.

Every licensed provider must also belong to one of four approved dispute resolution schemes, where compensation of up to $500,000 plus GST has been available since 18 July 2024. You can check who is registered on the Financial Service Providers Register, and Policywise's own details sit on the public disclosure page.

None of that reaches a chatbot. One lawyer working with the New Zealand insurance sector, a Chapman Tripp partner, points out that recommending a specific product crosses into regulated advice. That same lawyer found no AI chatbot currently holds a Financial Advice Provider licence.

The Financial Markets Authority is running a thematic review into AI in financial advice, with no findings published yet. A wrong AI answer has no licence behind it and no scheme to take it to.

The personal circumstances a chatbot can't weigh

A chatbot draws only on what you type into the prompt. It does not know your mortgage balance, your medical notes, who owns your policy, or whether you are an employee or a contractor. Those details change what cover you need and what it costs.

• A mortgage: Around 32% of New Zealand households have a mortgage on the primary residence, and for those households, median property debt reached $260,000 in the year ended June 2021. Mortgage protection insurance is sized to your number, not an average.
• Self-employment: In the June 2026 quarter, 354,000 people were self-employed with no employees. Employees get at least 10 days of paid sick leave a year after six months. A contractor does not get paid holidays. ACC covers injury caused by an accident, not illness or conditions related to ageing. That ACC income protection gap is what income protection for the self-employed is meant to close.
• A health history: In the 2024/25 New Zealand Health Survey, 14.3% of adults, about 619,000 people, reported high or very high psychological distress in the four weeks before the survey. As the investigation above shows, the same history is treated differently by different insurers.
• Family structure: One-parent families made up 16.5% of all New Zealand families in 2023, and 3.9% of households contained two or more families. Who owns a policy, and who a payout goes to, is a decision in those households rather than a default. Life insurance for families is often structured around this.

What happens to your cover if you act on a wrong AI answer?

Wrong information given to an insurer can void a policy, and AI tools give wrong answers about insurance in New Zealand. No New Zealand case has yet been reported of a consumer acting on a wrong AI answer and losing cover because of it, but what is documented is each half of that chain.

AI gets insurance questions wrong here

The Insurance and Financial Services Ombudsman has warned that AI can oversimplify complex policy wording, miss key exclusions, rely on overseas information that does not apply in New Zealand, and hallucinate previous cases.

In one example the scheme saw, a Google AI summary told a consumer that claim decisions are frequently overturned when people complain, with up to 80-90% succeeding if they persist. The Ombudsman calls that misleading. The Banking Ombudsman warns similarly that AI responses can make up laws, policies, or case examples that don't exist.

Wrong information given to an insurer has consequences

New Zealand law requires you to tell your insurer, when you first arrange a policy and at each renewal, about anything that could affect its decision to insure you. The remedy for non-disclosure is that the insurer can avoid or cancel the policy back to the beginning and treat it as if it never existed.

Current law does not distinguish an innocent omission from a deliberate one. In one case, a consumer described her husband's heart condition using the wrong term. FSCL accepted her non-disclosure was an innocent mistake, but still found the decline lawful because the insurer had been prevented from making an informed decision to grant cover.

Can you use AI and an insurance broker together?

Yes, and that is the sensible way to use both—AI to read and prepare, an adviser to decide and arrange. The Financial Markets Authority frames technology as automating parts of the process that take advisers a lot of time, freeing them up for the human conversations consumers value.

New Zealanders show a similar pattern. Comfortable with AI in a supporting role, markedly less so when it's making the decision. Research from the Financial Services Council, the financial services sector's industry body, found 42% of consumers are comfortable with financial services using AI overall, but only 35% are comfortable with AI making automated financial decisions.

Here’s a practical way to run it:

• Explaining, not recommending: Summarising a document you already hold is a fair use. Drafting the questions a good financial planner would ask you first is another good use.
• Checking what it tells you: The IFSO Scheme's instruction is to double-check any AI-generated information and advice against trusted, New Zealand-based sources.
• Bringing your own numbers: An adviser may ask about your personal situation, family, assets, debts including mortgages, income, expenses, existing policies, and estate plans. Having those ready shortens the conversation.
• Treating a comparison screen as a start, not the decision: Sorted, the government-funded money guidance service, suggests speaking to a financial adviser because insurance is complicated and can be set up in many ways to suit you.

When to stop asking AI and talk to a person

Talk to a person once the answer changes what you pay or whether you're covered. A health history and a disputed claim are the clearest triggers, but six situations below all turn on facts a chatbot cannot check.

• You have a health history, or you aren't sure what to disclose: If an insurer finds you haven't disclosed all your medical conditions or lifestyle choices, it can change the terms of your cover or cancel benefits, so ask your insurer or adviser if you are unsure. In one IFSO case, a knee-replacement claim was declined because medical records showed degenerative changes noted 15 years earlier.
• You are thinking about switching: Keep your old policy running until the new one is in force, or you could have a claim denied that might have been accepted under the original policy. Conditions you developed under your current cover may be treated as pre-existing by a new insurer.
• You have a claim already in dispute: If a complaint isn't resolved within two months of the provider first receiving it, or you and the insurer reach deadlock, it can go to your provider's dispute resolution scheme.
• You are working out how much cover to buy: Policy definitions can differ between providers. An adviser can look at your financial commitments and family situation and suggest how much you need.
• Your family or ownership situation is complicated: Whether a life payout forms part of your estate depends on who owns the policy. Under section 75A of the Life Insurance Act 1908, a policy on your own life expressed to benefit your spouse, de facto partner, or children creates a trust in their favour. This way, the money stays out of your estate while that trust is unperformed, unless the policy was taken out and the premiums paid with intent to defraud creditors, in which case creditors can recover an amount equal to the premiums paid, not the full payout.
• You own a business, or you have cover through work: For a key-person insurance policy the business owns, pays for, and benefits from, the payout is taxable business income and the premiums are deductible. Where an employer instead takes out a policy on behalf of its shareholders to buy out a shareholder's estate, the premiums are not deductible.

If any of these describe you, talk to a person rather than running another prompt. Talk to a Policywise adviser to see how the providers on Policywise's panel are likely to treat your situation, and what it will cost.

Helping you get advice you can hold someone to

Policywise is a 100% free service which tells you which health, life, and disability insurance provider best fits your needs. We offer fast, comprehensive, and easy-to-understand comparisons of all leading providers, and a simple summary clearly recommending which insurer is best for your situation.

Not all insurance policies are the same. Policywise can help you sort out the duds, avoid the lemons, understand the fine print and exclusions, and get the right insurance for you and your family.

We make the important decision of where to buy your insurance super easy. We’ll answer your questions, provide experienced advice and quotes, and manage all the back and forth throughout the application process. Taking out your cover through us means you'll have our lifetime support and claims advocacy, and we'll help you negotiate a positive outcome at claim time. We can also take care of lodging any claims on your behalf and back you up if the going gets tough.

Check out the reviews on our homepage for how other New Zealanders have found our service, because now is the time to get your personal, family, or business insurance sorted. Give your family or someone you love the most outstanding financial support possible. Book a 5-minute callback with Policywise today; our service is fast and free.