You Asked
We Answer
AI might generate the content, make a recommendation or automate the process, but your business can still be accountable for the outcome. The key is understanding where the risks sit, putting the right safeguards in place and making sure someone is responsible for checking what AI produces. AI is changing the way New Zealand businesses operate. Tasks that once took hours can now be completed in minutes, and tools that were once reserved for large organisations are increasingly accessible to smaller businesses.
The opportunity is significant, but so is the potential for things to go wrong.
An AI chatbot could provide incorrect pricing. An automated marketing campaign could publish misleading claims. An AI-generated article could contain inaccurate information that damages your credibility. And when that happens, customers aren’t interested in which algorithm made the mistake. They’re looking to the business they trusted.
The question isn’t whether businesses should be using AI. It’s how they can use it without losing control of the outcome.
Quick Links
The short answer is that responsibility depends on what went wrong, how the technology was being used and the legal obligations of everyone involved. AI doesn’t operate outside existing business laws simply because it’s a new technology.
In New Zealand, businesses using AI still need to comply with relevant legislation, including the Fair Trading Act 1986 and Privacy Act 2020. If AI-generated advertising misleads customers, the fact that a machine wrote the content doesn’t automatically remove the business’s obligations. Likewise, using a third-party AI platform doesn’t eliminate your responsibilities when handling personal information.
Depending on the circumstances, an AI provider, developer or other supplier may also have responsibilities. But from a customer’s perspective, the business they purchased from or interacted with is usually the first place they’ll turn.
Using AI to complete a task doesn’t automatically transfer accountability for that task.
For most businesses, the biggest risks aren’t futuristic scenarios involving autonomous machines. They’re everyday mistakes that can happen when AI-generated work goes unchecked.
AI makes it incredibly easy to generate advertising copy, social media posts, product descriptions and website content.
But speed doesn’t guarantee accuracy.
Imagine an AI tool creates an advertisement claiming your product delivers results that haven’t been proven, or promotes an offer with incorrect pricing. If that content goes live, your business could face customer complaints, reputational damage or legal consequences. That’s why AI-generated marketing still needs human review, particularly when it includes pricing, guarantees, product claims or financial information.
AI-powered chatbots can answer customer questions around the clock, reduce administrative workloads and improve response times.
But what happens when a chatbot confidently gives the wrong answer?
It might promise a refund that isn’t available, provide incorrect delivery information or misrepresent your terms and conditions. Without clear boundaries and escalation processes, a helpful automation can quickly become a customer service problem.
AI is increasingly being integrated into CRM systems, email marketing platforms, analytics tools and business workflows.
These integrations can save substantial time, but errors can also spread further when systems operate automatically. An incorrect customer segment, inaccurate recommendation or poorly configured workflow could trigger hundreds of actions before anyone notices.
The more independence you give an AI system, the more important it becomes to understand what it can do and how its decisions are monitored.
There’s a common assumption that because AI can complete a task, it should be trusted to manage the entire process. But generating an answer and understanding whether that answer is appropriate are two different things. AI can produce convincing information that is factually incorrect. It can miss context, misunderstand instructions and make recommendations based on incomplete information.
In digital marketing, for example, AI might identify an opportunity to increase advertising spend based on conversion data. But if the tracking has been configured incorrectly, that recommendation could lead to more money being invested in a poorly performing campaign. The technology isn’t necessarily the underlying problem. The issue is relying on its output without understanding the information behind it.
AI should make your team more effective, not remove the accountability that makes their work valuable.
Responsible AI adoption doesn’t mean slowing innovation or avoiding automation. It means building processes that allow you to take advantage of the technology while managing the risks.
There are five practical steps businesses can take.
Not every business process carries the same level of risk.
Using AI to brainstorm blog topics is very different from allowing it to automatically approve refunds, publish financial information or make decisions involving customer data.
Identify which tasks can be automated and which require human approval.
Establish clear review processes for customer-facing content, advertising, sensitive communications and business-critical decisions.
AI can prepare the work, but a person should remain accountable for approving important outputs.
Before connecting AI tools to your website, CRM or customer database, understand what information they access, where it’s processed and how it’s stored.
New Zealand’s Office of the Privacy Commissioner provides guidance on using AI in ways that protect personal information.
Businesses should consider privacy risks before introducing tools that process customer data.
Your staff may already be using AI tools to write emails, analyse information or generate content.
A straightforward internal AI policy can help establish which tools are approved, what information can be shared and what needs checking before it goes live.
It doesn’t need to be complicated. It just needs to be clear.
AI systems need ongoing monitoring, just like websites, advertising campaigns and other digital platforms.
Check outputs, review performance and make sure there is a clear process for correcting mistakes.
Where possible, maintain records of automated actions so problems can be investigated and resolved.
Absolutely.
As AI becomes more widely adopted, the ability to automate tasks will become less of a competitive advantage on its own. What will matter is how effectively businesses use that automation. Customers still value accuracy, reliability, transparency and trust. They want efficient service, but they also want confidence that the information they’re receiving is correct.
A business that combines AI efficiency with genuine human expertise can offer both. That might mean faster customer service with clear escalation to a real person, more efficient marketing campaigns supported by experienced strategists, or automated workflows that reduce administrative work without compromising quality. The most successful businesses won’t necessarily be those that automate the most. They’ll be those that know what to automate, what to check and when human judgement matters.
Yes. Businesses can remain responsible for content they publish, even if it was created using artificial intelligence. Relevant advertising, consumer protection, privacy and other laws can still apply.
Yes. AI-generated advertising must comply with applicable New Zealand laws and advertising standards, just like content written by a person. Businesses should review claims, pricing, offers and other factual information before publishing.
Not automatically. Responsibility depends on the circumstances, applicable laws and contractual terms. Businesses should review the conditions of the tools they use and understand their own obligations.
It depends on the risk involved. Low-risk, repetitive tasks may be suitable for automation, while decisions involving financial consequences, personal information or customer rights generally require stronger safeguards and human oversight.
Start with clear policies, appropriate access permissions, staff training, human review of important outputs and regular monitoring. Understanding the limitations of each AI tool is just as important as understanding its capabilities.
AI offers enormous opportunities for New Zealand businesses to improve efficiency, reduce costs and deliver better customer experiences. But the ability to automate something doesn’t automatically mean it should operate without oversight.
The businesses that get the most value from AI will be those that approach it strategically, understanding both what the technology can achieve and where it needs human support. Because when your AI gets something wrong, your customers probably won’t blame the technology.
They’ll blame the business behind it.
At NZ Digital, we help businesses make smarter decisions about the technology behind their digital growth.
From website development and digital marketing to AI opportunities and automation, we focus on solutions that deliver measurable value while keeping your business objectives front and centre.
If you’re exploring how AI could improve your operations, or want a clearer understanding of the opportunities and risks, our team can help you identify the right next steps.
Talk to NZ Digital and let’s explore how AI can work smarter for your business.
NZ Digital is an Auckland-based digital marketing agency specialising in lead generation, paid advertising, SEO, and digital strategy. If you’d like help understanding your performance or improving your results, get in touch for a strategy session.
SEO is one of the most important long-term investments in digital marketing, but choosing the right agency can be difficult when so many promise the same results. The difference often comes down to transparency, strategy and the questions you ask before signing.
Read More
Performance Max gives businesses a way to reach customers across Google’s entire advertising network from one campaign, but its success depends heavily on strong tracking, enough conversion data and clear commercial goals.
Read More
SEO is one of the most important long-term investments in digital marketing, but choosing the right agency can be difficult when so many promise the same results. The difference often comes down to transparency, strategy and the questions you ask before signing.
Read More
