Learning AI in Australia in 2026 costs anywhere from about A$300 a year to about A$60,000, and the price says almost nothing about the outcome. Self-taught with paid tools runs A$300–A$1,000 a year. A short course or microcredential is A$500–A$5,000. A twelve-week bootcamp is A$600–A$2,000 at the new online schools and A$10,000–A$20,000 at the established in-person ones. A university graduate certificate is roughly A$10,000–A$20,000; a masters, A$40,000–A$60,000 over two years. What you are buying at each price is different, and the expensive options are not buying more skill — they are buying credentials, cohorts and time.
Prices below are typical ranges at the time of writing; check the provider. Where we quote our own, they are the current pricing page.
What does it cost to teach yourself AI?
A$300 to A$1,000 a year, almost all of it in tool subscriptions and cloud credits — and the real cost is the discipline, which most people underestimate. A frontier-model subscription (Claude, ChatGPT or Gemini) is A$30–A$40 a month. Cloud credits to deploy things you build: A$10–A$50 a month if you are careful, more if you are not. The teaching material — documentation, the vendors' own courses, open lecture series — is free and, in 2026, very good.
The catch is completion. Self-paced online learning has had completion rates below fifteen percent for a decade, not because the material is bad but because nothing is waiting for the work. Self-teaching works for people who already ship things and just need the new stack. For a career changer it is the cheapest option and the one most likely to end with a folder of half-finished tutorials. How adults learn to code in 2026 is about why.
What do short courses and microcredentials cost?
A$500 to A$5,000, for two to eight weeks of structured material, sometimes with a badge at the end. TAFE and the universities' microcredential arms sit in this band, as do the vendors' own paid programmes and most of the online course platforms' "professional certificate" bundles. Some are FEE-HELP eligible when offered by an approved provider; most are not.
They are good for one thing: learning a specific, bounded skill — SQL, a cloud platform's AI services, prompt engineering for a role — with enough structure to finish. They are not a career change. Square 1's on-demand courses sit at the bottom of this band (A$49–A$249 at the time of writing) and are graded on the artefacts you produce rather than on a quiz; the AI for Project Managers, Python for AI and Evaluating AI Systems courses are typical of what fits in this price.
What does an AI bootcamp cost in Australia?
Two very different prices for the same twelve weeks. The established in-person and live-online bootcamps in Sydney and Melbourne charge A$10,000 to A$20,000 for a full-time or part-time immersive, and some offer payment plans or income-share arrangements. The newer online, AI-graded schools charge a tenth of that: Square 1's twelve-week bootcamps are A$590 to A$990 at founding rates (A$1,290 to A$1,990 standard), fifteen hours a week, six deployed projects, graded against a published rubric.
The gap is not a mistake and it is not a discount. The expensive bootcamps are paying for instructors in a room, a career-services team and a physical space; the cheap ones have replaced the instructor with an AI tutor that grades every line and replaced the room with a cohort online. Whether the first is worth ten times the second depends entirely on whether you need a human in the room to finish. Some people do. Is a twelve-week AI bootcamp enough to change careers is the honest treatment of what either kind can and cannot do.
What does a university AI qualification cost?
A graduate certificate in AI, data science or machine learning at an Australian university is typically A$10,000 to A$20,000 for four subjects over six months to a year part-time; a graduate diploma roughly double; a masters A$40,000 to A$60,000 for domestic full-fee students over two years. Most are FEE-HELP eligible, which defers the cost rather than reducing it. A small number of Commonwealth-supported postgraduate places exist in priority areas and bring the fee down sharply; they are competitive and worth asking about.
What the fee buys is the credential and the theory. For the doors that require a degree — government and defence panels, the banks' graduate programmes, skilled-visa qualification requirements, and research roles — nothing else substitutes. For an engineering role at a startup or a SaaS company, it is the most expensive route to a job that will be decided on your portfolio. Can you get an AI engineering job in Australia without a degree goes through which doors are which.
What do vendor certifications cost?
US$99 to US$150 for the entry-level exams (AWS Certified AI Practitioner, Microsoft's AI-900, Google's Generative AI Leader), US$150 to US$300 for associate and professional levels, plus whatever you spend on preparation. In our count of 315 AI job ads, a certification was mentioned in 7% — so this is a cheap door-opener for specific employers rather than a general requirement. Which cloud AI certification is worth it in Australia has the breakdown by employer type; the exam prep page has a free lane with the question bank.
What is the cost that nobody puts on the price list?
Time — and it is the one that decides the outcome. Fifteen hours a week for twelve weeks is 180 hours. Six months at seven hours a week is 180 hours too, and produces less, because momentum matters. The question to ask of any option is not "what does it cost" but "what does it cost per finished, deployed project", because in our count of the ads, shipped and production were the most-mentioned words (39%) and the price you paid to learn was mentioned in none.
The short answer, by situation
- Already a working programmer, self-disciplined: teach yourself; A$500 a year; build three things.
- Adjacent role, need structure to finish: an online bootcamp under A$2,000, or a short course for the specific gap.
- Need a human in the room to finish: an in-person bootcamp, A$10,000–A$20,000, and go in with your eyes open about what you are paying for.
- Need the credential for government, a graduate programme, a visa or research: a graduate certificate first; the masters only if the certificate confirms you want it.
- Everyone: the free skill checks before spending a dollar. Three minutes with a free student account, and they will tell you which of these you actually need.