Victoria has again positioned itself at the front of the national scramble for technology talent, with the state government signalling fresh support for the artificial intelligence and deeptech startups it hopes will anchor the next phase of its economy. The move, flagged through the Department of Jobs, Skills, Industry and Regions, is the latest sign that the states, rather than Canberra, are doing much of the heavy lifting when it comes to seeding early-stage founders working on hard science and frontier software.
The framing matters. Deeptech is the label the industry gives to companies built on genuine scientific or engineering breakthroughs rather than a clever app or a new marketing channel. Think quantum computing, advanced materials, robotics, medical devices, climate technology and the applied AI that increasingly sits underneath all of them. These businesses take longer to mature, chew through more capital before they earn a dollar, and carry more technical risk than a typical software startup. That combination has long made them harder to fund through private markets alone, which is exactly the gap governments keep trying to fill.
Why Victoria keeps leaning in
Victoria has spent the better part of a decade building the scaffolding for this kind of ambition. The state stood up LaunchVic, its dedicated startup agency, back in 2016, and has since layered on Breakthrough Victoria, a long-term investment vehicle designed to co-invest alongside private capital in research-heavy companies spun out of the state’s universities and medical research institutes. Melbourne’s cluster of universities, hospitals and research centres gives the state an unusually deep bench of the scientists and engineers who tend to start deeptech companies, and successive governments have treated that concentration as a competitive asset worth protecting.
The logic behind the latest backing is straightforward. Artificial intelligence has moved from a niche research field to a general-purpose technology that touches manufacturing, health, agriculture, defence and financial services. States that host the companies building and applying that technology stand to capture the high-value jobs, the intellectual property and the follow-on investment that come with it. States that miss the window risk watching their best founders decamp to Sydney, Singapore or San Francisco, taking their research and their future headcount with them.
Two ways to read the strategy
Supporters of this kind of intervention argue that it is precisely the role of government to underwrite risk that private investors will not take on their own. Deeptech timelines can stretch to a decade, which sits awkwardly with the shorter horizons of many venture funds. A patient public co-investor, the argument runs, can keep a promising company in the country long enough to prove itself, at which point private capital is far more willing to follow. The public payoff comes later, in the form of new industries, sovereign capability in areas like semiconductors or quantum, and a pipeline of skilled workers who might otherwise never have had a local employer to join.
The sceptics are not hard to find either. Critics of government-led startup programs point out that picking winners is difficult even for professional investors who do it full time, and that the incentives around public money can push officials towards safe, announceable projects rather than genuinely risky bets. There is also a persistent question about duplication, with the Commonwealth, the states and a growing list of agencies all running their own grants, funds and accelerators. Founders frequently complain that the resulting patchwork is confusing to navigate and that the paperwork attached to public money can slow them down at exactly the moment they need to move fast. The honest answer is that both things can be true at once: the support is genuinely useful, and the system around it could be simpler.
The national stakes
What gives this announcement weight beyond Victoria’s borders is the contest it feeds. New South Wales has its own investment vehicles and a heavy concentration of the country’s venture capital. Queensland has leaned into advanced manufacturing and clean energy. South Australia has built a defence and space cluster around Adelaide, and Western Australia is pushing hard on the intersection of AI, mining and robotics. Every state is now, in effect, running an industry policy, and every one of them is chasing a similar pool of founders, researchers and investors.
That competition is mostly healthy for the country. It drives up the total amount of capital and support available to Australian founders, and it gives entrepreneurs real choice about where to build. The risk is fragmentation. Australia is a mid-sized economy trying to compete against far larger rivals for scarce technical talent, and eight separate approaches to AI and deeptech can pull in different directions. A founder weighing up whether to stay in Australia at all is comparing the whole national offer against Silicon Valley or Europe, not Melbourne against Sydney. The states that recognise this and design their programs to complement rather than cannibalise each other are likely to get the best long-run return.
There is a skills dimension too. AI and deeptech companies need machine learning engineers, research scientists, hardware specialists and the commercial operators who can turn a prototype into a product. Australia does not produce enough of those people, and the ones it does are highly mobile. Backing startups without also widening the talent pipeline, through universities, migration settings and retraining, risks funding companies that then cannot hire. The most durable state strategies treat capital and talent as two halves of the same problem.
What comes next
The practical test of any of this is not the announcement but the follow-through: how much money actually reaches founders, how quickly, and whether the companies it backs go on to raise private rounds, hire locally and stay in the country. Those outcomes take years to show up, which makes programs like this politically awkward, because the credit and the results rarely land in the same electoral cycle. The states that stay the course, keep their support simple to access, and resist the urge to chase headlines over substance will be the ones still standing when the current wave of AI enthusiasm settles into something more durable.
For Australia’s founders, though, the direction of travel is clear enough. State governments have decided that AI and deeptech are worth competing over, and that competition is putting more support within reach than at almost any point in the past decade. The task now is to make sure the country turns that appetite into companies that endure.
Sources: Department of Jobs, Skills, Industry and Regions (Victoria).
