Country guide · New Zealand

EU AI Act for NZ businesses: does it apply to you?

Last updated: 2 November 2026

Key point: The EU AI Act doesn't stop at Europe's borders. If people in the EU use your chatbot, talk to your AI phone agent or see your AI-generated images, Article 50 can apply to your New Zealand business, with no European office needed (Article 2).

New Zealand has chosen a light-touch path on AI, with no AI Act of its own. That makes the EU's rules easy to overlook from here. For Kiwi exporters, tourism operators and SaaS teams with European customers, Article 50 has applied since 2 August 2026.

Scan for free

Does the EU AI Act apply to New Zealand businesses?

It can. Article 2(1)(c) of the AI Act covers “providers and deployers of AI systems that have their place of establishment or are located in a third country, where the output produced by the AI system is used in the Union”. New Zealand is a third country, so the question isn't where your company is registered. It is where your AI's output ends up.

In practice there are three routes in. You sell an AI system or an AI feature to customers in the EU. You have an establishment in the EU, such as a European subsidiary. Or people in the EU use the output of AI you run, such as answers from your website chatbot. For most New Zealand websites, the third route is the one that matters.

The test is where the output is used, not the visitor's nationality. A business that serves only New Zealanders, and doesn't market to Europe, is unlikely to be in scope. This is general information, not legal advice.

Which NZ businesses are most exposed

SaaS with EU subscribers

A Wellington practice-management platform with subscribers in Germany, whose AI assistant drafts client replies. As the provider of that feature, it owns the Article 50(1) duty to tell users they are working with AI.

Exporters selling online

A Nelson merino knitwear label shipping to France and the Netherlands, with an AI chat widget on its online store. European shoppers chatting with it are using the output in the EU.

Tourism operators

A Queenstown adventure-tourism company whose booking chatbot answers travellers while they plan from Germany, or a Christchurch campervan-hire firm with an AI phone agent taking calls from Europe.

Education providers

An Auckland English-language school whose enrolment chatbot talks to prospective students in Spain and Italy before they apply.

Agencies and studios

A Tauranga marketing studio making realistic AI-generated video of people for a European client. That can be a deep fake under Article 50(4), which needs a disclosure.

Probably not in scope

A Hamilton physio clinic with an AI booking assistant, serving Waikato patients only. No EU market, and no output used in the EU.

These are illustrative examples, not real businesses, and not legal advice. Exporters may also find the ecommerce guide useful.

Article 50 in plain English

Article 50 doesn't ban anything. It makes sure people know when they are dealing with AI. Put simply, it asks three questions of a New Zealand website with European visitors:

→Does your bot say it's a bot? Under Article 50(1), a chatbot or voice agent must be designed so people are told they are talking to AI, unless that is obvious. The duty falls on the provider, so if you use a vendor's tool, check its AI notice is turned on and actually shows. Our chatbot disclosure guide has wording you can paste in.
→Is AI-generated content marked? Article 50(2) asks providers of generative AI to mark their output in a machine-readable way. Older systems have until 2 December 2026 (Article 111(4)). More in AI content labelling.
→Are deepfakes labelled? Article 50(4) asks whoever publishes a realistic AI-generated or manipulated image, video or audio of people, places or events to say so. See deepfake disclosure.

The notice has to be clear and distinguishable, and shown at the latest at the first interaction or exposure (Article 50(5)). Article 50 has applied since 2 August 2026 (Article 113), and breaches can lead to fines of up to €15 million or 3% of worldwide annual turnover, with the lower figure for SMEs (Article 99). For the full text, read our Article 50 guide.

AI regulation in New Zealand: what applies at home

New Zealand has no AI-specific legislation. The government has chosen to manage AI through the laws it already has, and to encourage adoption with voluntary guidance.

→Strategy for AI, July 2025. The government launched New Zealand's first AI strategy in July 2025. The strategy, “Investing with confidence”, describes a light-touch, principles-based approach that relies on existing legislation such as privacy, consumer protection and human rights law, rather than a standalone AI Act.
→Responsible AI Guidance for Businesses. MBIE's guidance, published alongside the strategy (MBIE), is voluntary. It sets out good practice and points businesses to the existing laws that already apply to AI.
→Privacy Act 2020. The Office of the Privacy Commissioner says the Privacy Act applies to anyone using AI tools in New Zealand, and explains how each Information Privacy Principle applies (AI and the IPPs, PDF).

None of this switches the EU rules off. A New Zealand business can meet every local expectation and still owe Article 50 duties to its European users. Across the Tasman the picture is similar but not the same: see the guide for Australian businesses.

Built in New Zealand

Article50.io is built here in New Zealand by Maynard, a solo founder. Every question comes straight to me, and I reply within 24 hours. More about Article50.io.

Sources

This page is general information, not legal advice. A clean result is not a compliance clearance, and this is not legal advice.

Frequently Asked Questions

Does the EU AI Act apply to New Zealand companies?

It can. Article 2 extends the AI Act to providers and deployers outside the EU when the output of their AI system is used in the EU. A New Zealand company with no European office can be in scope if people in Europe use its chatbot, AI voice agent or AI-generated images and video.

Does New Zealand have an AI law?

Not a dedicated one. New Zealand's Strategy for AI, released in July 2025, takes a light-touch, principles-based approach that relies on existing laws such as privacy, consumer protection and human rights law. MBIE's Responsible AI Guidance for Businesses is voluntary. The EU AI Act sits outside that and can still apply.

Does the Privacy Act 2020 apply to AI chatbots?

Where personal information is involved, yes. The Office of the Privacy Commissioner says the Privacy Act applies to the use of AI tools in New Zealand, and its guidance maps the Information Privacy Principles onto AI. That is a separate question from Article 50, which is about telling people they are dealing with AI.

When did Article 50 start to apply?

On 2 August 2026, under Article 113 of the AI Act. Providers of generative AI systems already on the market before that date have until 2 December 2026 to add Article 50(2) machine-readable marking. The chatbot and deepfake disclosure duties already apply to New Zealand businesses that are in scope.

What are the fines for a New Zealand business?

A New Zealand business in scope faces the same ceiling as an EU one. Article 99(4) sets fines for Article 50 breaches of up to €15 million or 3% of total worldwide annual turnover, whichever is higher. For SMEs, including start-ups, Article 99(6) applies whichever of the two amounts is lower.

Check your site automatically

Run the free scan on your homepage or booking page. It opens the page in a real browser and shows your single most severe Article 50 finding in under a minute, no signup needed. A clean result is not a compliance clearance, and this is not legal advice.