ShareFacebookX (formerly Twitter)Artificial intelligence (AI) is being touted as the next big thing in economic growth. It promises to transform industries, boost efficiency and make life in general easier. In Australia, businesses and policymakers are devoting time and money to AI, hoping it will drive productivity. Yet, despite the hype, national productivity growth remains dismal.So, what’s going on? If AI is so revolutionary, why aren’t we seeing the benefits? The reality is that big technological innovations take time to pay off. Think about the introduction of computers in the late twentieth century. They completely changed the way we work, but their effect wasn’t immediate. It took time and required new business processes, and workforce reorganisation, to say nothing of training, before productivity gains started to materialise.AI is likely to follow the same path — it’s not an instant fix, but it has the potential to be a game changer if implemented correctly.Where AI can make a differenceAustralia’s economy is now built on services, which make up80 per cent of the nation’s economic activity. While industries like mining and manufacturing have been using automation to drive productivity for years, many service sectors have not been as quick to adapt. AI could help change that, offering new ways to lift efficiency, support better decision-making, and transform how work is done in areas like healthcare, education and retail.But there’s a catch. AI can’t just be about replacing workers with machines. True productivity gains that boost living standards will come from using AI toenhancehuman work — making processes smarter, helping people work more effectively and creating new opportunities rather than eliminating jobs. If businesses only see AI as a cost-cutting tool, society risks missing out on its real benefits.The roadblocks to AI adoptionEven with all its potential, AI adoption will be uneven. Large corporations with more resources will be early adopters, as we have already seen inmajor banks using AI for fraud detectionandmining companies employing autonomous trucksto boost efficiency. However, for small and medium-sized businesses, the challenge is less about cost — given the accessibility of tools like ChatGPT and Copilot — and more about relevance and confidence. Many smaller firms may not see how AI fits into their business, lack the technical understanding to use it effectively, or simply feel uncertain about experimenting with new technology.Want the best of Religion & Ethics delivered to your mailbox?Sign up for our weekly newsletter.Your information is being handled in accordance with theABC Privacy Collection Statement.Email addressSubscribeThere’s also the issue of regulation and risk. Industries like healthcare, where strict compliance is required, or law, where AI adoption faces ethical and professional constraints, are less adaptable compared to sectors like logistics, where AI-driven route optimisation is already improving operations. In healthcare, accuracy is critical — while an error in a legal case summary is inconvenient, a misdiagnosis could have serious consequences. This makes AI adoption not just a regulatory challenge but also a risk management issue, because firms must balance potential benefits against the risk of mistakes.A big part of the challenge is that while AI tools like ChatGPT are plug-and-play and widely accessible, integrating AI meaningfully into business operations often requires more than just adoption. Businesses may need to rethink their processes, build new expertise and retrain workers to fully unlock its potential. Without these efforts, AI may be used in limited ways rather than transforming how businesses operate and contribute to long-term economic growth.The big ethical questionEven if AI boosts productivity, not everyone will benefit equally. High-skilled workers in fields like finance and tech may see their efficiency soar, but what about those in jobs more vulnerable to automation — call centre employees, for instance, or factory staff? For them, AI could mean job losses rather than opportunities.Take theRobodebt scandalas an example. The automated welfare debt recovery system falsely accused thousands of vulnerable Australians of owing money, causing immense hardship. It was a clear case of automation being used irresponsibly and without proper safeguards, leading to devastating consequences.Without the right guardrails, similar mistakes could happen with AI on a larger scale. That’s why businesses and policymakers must ensure automation and AI-driven decisions are transparent, fair and accountable.Making AI work for everyoneFor AI to deliver real, widespread benefits to living standards, we need to develop a strategy. Workers will need opportunities to develop new skills so they can work alongside AI rather than be replaced by it. Small businesses will require support to adopt AI effectively, not just big corporations.At the same time, regulations must ensure AI is used responsibly, preventing harm and ensuring fair outcomes. Above all, AI investment should be focused on improving how people and technology work together — not just cutting jobs.Luke Hartiganis a Lecturer in the School of Economics at the University of Sydney.Stella Huangfuis an Associate Professor in the School of Economics at the University of Sydney.Posted17 Mar 202517 Mar 2025Mon 17 Mar 2025 at 1:33amShareFacebookX (formerly Twitter)Resisting AI hyper-optimism: The Paris AI summit should concern us allWhy we shouldn’t want to be the pets of super-intelligent computersWhat we can learn from the failure of RobodebtWhy automating good behaviour in the workplace could end up making you a worse personArtificial Intelligence,EthicsBack to top