What will AI mean for our work, our businesses and everyday life? These are the questions I’m working through.
Every Australian business should have a practical option to use capable models served here under Australian control. The test is credible legal and operational continuity, while retaining international partnerships and the freedom to use overseas services.
The sign and distribution of Australia’s full net AI benefit remain unresolved here. Domestic output is useful evidence, but it is not the same as income retained by Australians or welfare after transition and environmental costs.
Model access is only one part of useful adoption. Organisations also need suitable workflows, reliable review, clear responsibilities and a reason to change. Some delay may be avoidable; some reflects genuine costs and constraints.
Substantial public investment can be justified when it secures a clear public benefit. Contributions through taxes or utility bills should be explicit and justified, with a fair developer contribution and practical capability for businesses.
Early-career entry routes warrant attention. Recent evidence suggests pressure on junior hiring in some exposed settings, alongside opportunities for AI to help beginners. It does not establish that the Australian career ladder has disappeared.
It could help, but a faster task is several steps away from a cheaper household bill. The size, timing and distribution of gains depend on implementation, competition, wages, prices and supply constraints.
Students need to learn how to use AI and how to judge the work it produces. The evidence favours attention to instructional design and independent competence. It does not establish that unaided practice is obsolete or that all AI assistance harms learning.
Imported AI can create Australian gains while leaving frontier dependence intact. AI-assisted research could compound the leaders’ advantage, but the size and persistence of that effect remain unsettled. Useful adoption, local continuity and assured frontier access need separate tests.
AI can shift work from the sender to the recipient and the queue. Some extra demand also represents useful access we previously failed to deliver. Count worthwhile resolutions and total effort; current evidence cannot tell us what proportion of the new demand passes that test.