Artificial Intelligence · Analysis

Australia's New AI Laws: What Small Businesses Need to Know Before 2027

Australia is preparing tougher AI safety rules. Find out what's proposed for 2027, what laws apply today and what small businesses should do now.

Harbour Creative TeamHarbour Creative Team

6 min read

Keep this story in your reading list
Australia's New AI Laws: What Small Businesses Need to Know Before 2027
Share article
0%
Ways to read

Australia is preparing stronger rules for artificial intelligence. Here's what's changing, what's already covered by law, and what business owners can do without waiting for Parliament.

A few years ago, using AI in a small business might have meant experimenting with a chatbot.

Today, the same business could use AI to answer customer enquiries, write advertising copy, process documents, screen applications and automate tasks that previously required several employees.

On 8 October 2026, the Australian Government outlined its direction on advanced AI safety and future regulation.

For the average Australian business owner, the practical question is straightforward:

What does this actually mean for my business?

While the announcement signals a major shift in national technology policy, it is crucial to separate what is currently law from what is proposed for the future.

What Australia Announced on 8 October 2026

Speaking at the Sydney Trust and Safety Festival, Assistant Minister Andrew Charlton outlined the government's approach to securing the benefits of artificial intelligence while mitigating its risks.

The central component of the government's direction involves stronger safety expectations for the developers of advanced AI systems. The focus is squarely on "frontier" AI models, those powerful foundational systems that carry systemic risks.

To manage these risks, the government is developing mandatory guardrails designed specifically for the developers and deployers of high risk systems, and is establishing national standards to provide certainty for industry.

However, the distinction between a multinational technology firm building a frontier AI model and a local Australian company using an AI tool is critical.

Australia has not introduced a universal new AI licence for small businesses using everyday AI tools.

As of publication, ordinary businesses adopting commercial AI products for standard operations do not face an immediate new licensing regime. But that does not mean AI use is entirely unregulated.

What AI Laws Already Apply to Australian Businesses?

The Regulatory Landscape

Understanding what applies today versus what is being developed for tomorrow.

Category 01

Already Applicable
  • Australian Consumer Law (ACL)
  • Privacy Act 1988 (Cth)
  • Copyright Act 1968
  • Fair Work & Anti-discrimination

Applies depending on the business size and specific activities.

Category 02

Proposed / Developing
  • Mandatory guardrails for high-risk AI
  • Advanced AI safety standards
  • AI-specific consumer protections
  • Privacy automated decision reforms

Subject to parliamentary processes and public consultation.

Category 03

Recommended
  • National AI Centre's 6 Practices
  • Voluntary AI Safety Standard
  • Transparent AI disclosure rules
  • Internal acceptable use policies

Good business practice, not currently enforceable as separate laws.

While we wait for AI specific legislation, businesses must remember that existing Australian laws apply to AI activities just as they apply to human activities. Four areas require particular attention.

Australian Consumer Law

The Australian Consumer Law (ACL) prohibits misleading or deceptive conduct. If a business uses an AI system to generate marketing content that makes false claims about a product, the business remains responsible. Similarly, publishing AI generated testimonials or failing to ensure chatbots provide accurate pricing information can breach consumer protections.

Privacy and Customer Information

When a business uses AI to process customer data, the Privacy Act 1988 (Cth) and the Australian Privacy Principles may apply. Uploading sensitive customer information to a third party AI service without appropriate consent or security measures creates significant risk. Note that while the Privacy Act currently applies primarily to businesses with an annual turnover exceeding $3 million, proposed reforms may change these exemptions.

Employment and Workplace Responsibilities

Businesses increasingly use AI to screen recruitment applications or monitor workplace productivity. However, anti discrimination laws and the Fair Work Act remain in force. If an AI tool filters candidates in a way that unlawfully discriminates based on age, gender or race, the business using the tool can be held liable. Human oversight in consequential employment decisions is strongly advised.

Copyright and Intellectual Property

Generating commercial content using AI requires care. The Copyright Act 1968 protects original works. If an AI tool produces material that substantially reproduces an existing copyright work, using that material commercially could constitute infringement. Businesses should review the licensing terms of their AI providers and carefully check the outputs before publishing.

Not All Business AI Comes With the Same Risks

Where AI Appears in Small Businesses

Artificial intelligence is no longer restricted to specialised tech firms. It is increasingly embedded across everyday operations.

Marketing

  • Copywriting
  • Ad targeting
  • SEO optimisation

Customer Service

  • Chatbots
  • Email triaging
  • FAQ answers

Recruitment

  • CV screening
  • Interview summaries
  • Skill matching

Operations

  • Inventory forecasting
  • Routing
  • Pricing automation

Administration

  • Data entry
  • Meeting transcripts
  • Contract summarisation

To understand practical compliance, consider how risk varies depending on the application.

Example A: A Sydney café The café uses AI to draft captions, suggest seasonal promotions and summarise customer feedback. The main risks involve inaccurate claims, brand reputation damage and potential mishandling of customer data if names are uploaded. Because the consequences of an error are relatively minor, the required safeguards are straightforward. A staff member checking the copy before posting is usually sufficient.

Example B: An Australian recruitment company The company uses AI to screen applicants and assist with shortlisting for executive roles. Here, the consequences of an error or bias are severe, directly impacting an individual's livelihood and exposing the company to discrimination claims. This higher consequence use requires greater attention to accuracy, fairness testing and strict human oversight.

How Risky Is Your Business's AI Use?

Not all AI applications carry the same consequences. Explore four common business scenarios.

AI Marketing and Content

Lower Risk
Potential Concerns
  • Inaccurate advertising claims
  • Copyright or licensing issues
  • Brand reputation
Recommended Safeguard

Have a person check facts, permissions and final copy before publishing.

These examples illustrate potential risks. They are not official Australian Government classifications or personalised legal assessments.

Six Things Australian Small Businesses Can Do Today

Instead of waiting for new legislation, businesses can adopt voluntary best practices immediately. The National AI Centre has published guidance to help Australian organisations build safe and responsible AI foundations.

Six Essential AI Practices

Based on the Australian Government's National AI Centre guidance. Implementing these voluntary practices helps ensure safe and responsible AI adoption.

01

Establish accountability

Define who is responsible for the AI system and its impacts within your business.

02

Understand impacts

Assess how the AI system could affect people, including customers and employees.

03

Identify and assess risks

Evaluate potential harms, such as bias, inaccuracy, or privacy breaches.

04

Be transparent

Let people know when they are interacting with AI or when AI is making decisions about them.

05

Test and monitor

Regularly check that the AI system works as intended and remains secure over time.

06

Maintain human oversight

Ensure humans remain in the loop, particularly for consequential decisions.

Information adapted from the National AI Centre. This does not constitute an official government endorsement.

These practices help build trust with customers and establish internal processes that will prove valuable if formal regulations are introduced.

What Could Change Before 2027?

Businesses should monitor several developing areas:

  • Proposed advanced AI safety legislation: The government continues to consult on mandatory guardrails for high risk AI development and deployment in Australia.
  • Privacy and automated decision making developments: Separate reforms to the Privacy Act are progressing, which may include new requirements for transparency when businesses use personal information in automated decisions.
  • Industry specific rules: Sectors such as healthcare, finance and education may introduce specific AI use guidelines through their respective regulators.

Should Australian Businesses Stop Using AI?

Responsible adoption does not mean avoiding AI entirely. It means understanding the context of your specific use case.

Lower consequence uses, such as drafting internal meeting summaries or generating initial marketing concepts, offer genuine productivity benefits with manageable risks. Higher consequence uses involving personal information, financial decisions or employment require significantly more caution and human intervention.

The Bottom Line

The most practical thing a business can do today is understand exactly which AI systems it uses, what information those systems can access, what decisions they influence and who remains responsible for their outputs.

Businesses do not need to predict every future law. They do need to understand what their technology is doing today.


Frequently Asked Questions

Does Australia already have AI laws? Yes. While Australia is developing AI specific safety legislation, existing laws, including the Australian Consumer Law, the Privacy Act, and copyright protections, already apply to activities conducted using AI.

Will Australian small businesses need an AI licence? The government has not proposed a universal AI licence for all businesses. Current proposals focus on mandatory guardrails for developers and deployers of high risk, advanced AI systems.

Can Australian businesses use ChatGPT for customer information? Businesses must be careful. Uploading personal customer data to external AI services can breach privacy obligations. You must ensure you have appropriate consent, security measures and understand how the provider uses your data.

When will Australia's new AI laws take effect? The government is currently designing the legislative framework. While consultations and standards are developing now, parliamentary passage and commencement dates for specific new AI Acts remain subject to the legislative process.


Research and Sources

Information current as of 10 October 2026. This article provides general information and does not constitute legal advice.

About this article

This article provides analysis and interpretation of available information. Sources and supporting material, where applicable, are linked within the article.

Found an error? Contact the editorial desk. See our editorial standards and corrections policy.

From the Harbour Creative Blogs archive

Keep exploring this story

Related reporting, people and places from our published coverage. These are editorial connections, not a popularity ranking.