Skip to content

Start with one repetitive job, a tool you already pay for and a one page policy. Here is how Australian small businesses can put AI to work safely.

AI & Data Science8 min read

The best place for a small business to start with AI is one repetitive job that eats hours every week, using an AI tool it already pays for, with a short written rule about what information goes in. Custom AI comes later, and only if the everyday tools cannot do the job.

Most Australian small businesses have not started yet, which is an opportunity for those that do. This guide covers where small businesses stand, what they use AI for, the privacy rules that apply, and how to pick a first task. When a process is specific enough to need its own solution, that is the work our AI and data science team does.

How many small businesses use AI

It depends on who is asking. The Australian Bureau of Statistics found that around 12 percent of Australian businesses used AI in their workplace in 2024-25, rising from about 11 percent of small and micro businesses to 35 percent of large ones. The National AI Centre's monthly survey of small and medium enterprises found that 43 percent reported some level of AI adoption between December 2025 and February 2026. The surveys ask different questions of different groups, but they agree on the shape: most small businesses are not yet using AI in a meaningful way.

Australian businesses using AI, by size, 2024-25Share of Australian businesses that reported using AI in their workplace in the 2024-25 financial year.
  • Large businesses35%
  • Medium businesses22%
  • Small and micro businessesabout 11%
  • All businessesabout 12%

Source: Australian Bureau of Statistics, Characteristics of Australian Business 2024-25, released 25 June 2026.

Size is not the whole story. The same ABS survey found that small businesses that had introduced or improved a product, service or process during the year used AI at 19 percent, almost five times the rate of those that had not. AI tends to follow businesses that are already looking for better ways to work.

What small businesses use AI for

Among the small and medium businesses using or planning to use AI, the National AI Centre found the most common uses were content generation and data analytics, followed by cybersecurity and threat detection. These are productivity tasks with a person checking the result, which is the right place to start.

What Australian SMEs use AI forShare of small and medium businesses using or planning to use AI that apply it to each task, December 2025 to February 2026.
  • Content generation54%
  • Data analytics54%
  • Cybersecurity and threat detection48%

Source: National AI Centre SME AI Pulse, December 2025 to February 2026, carried out by Fifth Quadrant.

In a business of five to two hundred people, that usually looks like this:

  • Drafting replies to common enquiries, quotes and proposals from your own templates
  • Turning meeting and call notes into a summary and a list of actions
  • Reading supplier invoices and receipts into the accounting system for someone to approve
  • Answering staff questions from your policies, procedures and price lists
  • A first pass over sales, job or stock data to spot trends worth a closer look
  • Checking forms and documents for missing information before they go out

Why most small businesses have not started

The National AI Centre asked. Around 65 percent of businesses not adopting AI said they distrust AI decision making or want to keep human control of their processes, 54 percent said AI is not relevant to their business, and 19 percent of SMEs said they simply do not know how to use it.

All three are reasonable, and none is a reason to stay out. Keeping a person in control is how AI should be used in a small business anyway: the tool drafts and a person decides. Relevance becomes obvious once you look at where the hours go each week rather than at what AI can do in general. And not knowing how is solved by starting small, with one task and one tool.

Start with the tools you already pay for

Most software small businesses already use now includes AI features, from Microsoft 365 and Google Workspace to accounting, CRM and help desk systems. Starting there avoids a new contract, keeps your information inside a service you have already assessed, and means staff learn AI inside tools they know.

Pick one task, give it a month, and measure it: how long the task took before, how long it takes now, and how often the AI's draft needed correcting. If the numbers are good, roll it out to the rest of the team. If they are not, you have lost a month and learned what your information needs before AI can use it.

Keep customer and staff information safe

Most businesses with an annual turnover of $3 million or less are exempt from the Privacy Act, although health service providers and businesses that trade in personal information are covered whatever their size. A draft bill released in August 2026 proposes narrowing the exemption further, but it is not law. Even where the Act does not apply, customers expect their information to be handled carefully, and larger clients often write privacy obligations into their contracts.

The Office of the Australian Information Commissioner's guidance on commercially available AI products is clear on the main points. Privacy obligations apply to what goes into an AI tool and to what comes out of it. Public facing tools such as chatbots should be identified as AI. And as best practice, organisations should not enter personal information, particularly sensitive information, into publicly available generative AI tools.

A one page AI policy covers most of what a small business needs:

  • Which AI tools staff may use, and on which accounts
  • What information must never go into them, such as customer details, staff records and anything confidential
  • Who checks AI output before it reaches a customer, a supplier or a regulator
  • How customers will know when they are dealing with AI
  • Who to tell if something goes wrong

For a fuller checklist, the National AI Centre's Guidance for AI adoption: foundations is written for organisations at the start of their AI use.

When custom AI is worth it

Off the shelf tools cover general tasks well. Custom AI earns its cost when a process is specific to your business, involves your own systems and records, and happens often enough that small savings add up. An assistant that answers customers from your own product data and order history, a tool that classifies the documents your industry handles, or an agent that prepares records in your job management system are typical examples.

The same rules apply as for any AI: start with one process, keep a person approving the results, and measure before you widen it. Our post on agentic AI and generative AI explains the difference between an assistant that drafts and an agent that acts.

What we would do

List the five tasks that take the most staff hours each week and pick the one that is most repetitive and easiest to check. Try the AI features in the software you already use on that task for a month, with the one page policy in place from the first day. Measure the result, then decide on the next task. Most small businesses never need more than this, and the ones that do will know exactly which process to build for, which is where our AI and data science work usually begins.

Questions about AI for small business

  • Small businesses most often use AI to draft emails, quotes and proposals, summarise meetings, read invoices into accounting systems, answer staff questions from their own policies and analyse sales data. The best first use is one repetitive task that takes hours each week and is easy for a person to check.

  • The ABS found that around 11 percent of small and micro businesses used AI in their workplace in 2024-25, against 35 percent of large businesses. The National AI Centre's survey found that 43 percent of small and medium enterprises reported some level of AI adoption between December 2025 and February 2026.

  • Usually the one built into software you already pay for, such as Microsoft 365, Google Workspace or your accounting and CRM systems. It keeps your information inside a service you have already assessed and needs no new contract. Choose by the task you want to improve, not by the tool.

  • The Office of the Australian Information Commissioner recommends that organisations do not enter personal information, particularly sensitive information, into publicly available generative AI tools. Use business versions of AI tools with clear data terms, and keep customer and staff details out of free public ones.

  • Most businesses with an annual turnover of $3 million or less are exempt, but health service providers and businesses that trade in personal information are covered regardless of size. Larger clients often require privacy compliance in their contracts, so it is wise to follow the rules either way.

All insights