Why have many businesses looked at AI and said no?
Across the thirteen AI applications the National AI Centre tracks, the most common answer is the same one every time: aware of it, not likely to adopt. That is not ignorance, and it is not unreasonable.
Being right to doubt it isn't the same as knowing.
That's the gap this fund exists to close.
It's an evidence problem, not an AI problem.
Figures: National AI Centre AI Adoption Tracker, surveyed by Fifth Quadrant. December 2025 to February 2026 quarter, published May 2026. Minimum 400 Australian SME owners and decision-makers per monthly wave, weighted by industry, state and employee size to the national distribution of businesses. National figures across Australian small and medium businesses, not a manufacturing-specific sample. ai.gov.au
What this could look like
Five examples of what testing properly looks like in a manufacturing operation. Open any to read it.
What would it be worth to know a machine's about to fail, before it does? The study would test whether sensor data and AI-based diagnostics can genuinely predict failure ahead of your current maintenance approach, using your own equipment history as the benchmark, not a vendor's case study.
Manual inspection catches most defects. It's the ones it misses that cost you, in rework, returns, and reputation. Worth testing whether automated or AI-assisted inspection closes that gap, and putting a real number on what the gap is actually costing today.
There's probably one task in your operations nobody wants, and it's costing you every time someone calls in sick or walks. A feasibility study puts a real number on whether automating it stacks up, cost, retraining, and payback, not the number a vendor quotes you.
If someone asked you right now where job 47 is, could you answer without walking the floor? Digital job and materials tracking is worth testing properly, benchmarked against how you actually track things today, not how the software demo makes it look.
Switching material or process is a one-way bet if you get it wrong. A feasibility study benchmarks the alternative against what you use now on cost, quality and supply reliability, before you commit an entire production run to finding out the hard way.
It pays for outside expertise
The fund exists to bring an external assessment into your business. It funds an external consultant or manufacturing service provider to run the feasibility work, not internal effort you could have spent anyway.
At least 75% of the project budget must go to that external provider. Internal wages are capped at 25%, and a business case you've written in-house is explicitly ineligible. Your application names who the provider is and what they'll do.
You end up with an evidence-based report on technical feasibility and commercial viability, with a recommended adoption pathway. It's also your business case for Stage 2, which opens in early 2027 with grants of up to $5 million.
Is this you?
Some of these are firmer than others, and the turnover test and the lower end of the employee band are the two that catch people out. If you're not certain, a 30 minute call is the fastest way to find out whether it's worth your time.
Assessment before adoption is what we do anyway.
Fractional CAIO, strategic advisory. IMAF Stage 1 just happens to fund it. We're not here for the grant, and we'll still be here in November.
We'll take a call with anyone eligible. We can only deliver a handful of these this round, so early matters.
Start with a 30 minute call.
If it's a fit, you'll have a draft proposal within two business days, ready to attach to your application.
