Recently, there has been a lot of talk about productivity, especially among politicians. And justification: Australia's labour productivity growth is at 60 years' low.
To address this, Prime Minister Anthony Albanese convened a productivity roundtable next month. This coincides with the release of a preliminary report from the Productivity Committee, which is considering five reform pillars. One of these is the role of data and digital technologies, including artificial intelligence (AI).
This is music to the ears of the tech and business sectors that are enthusiastically promoting the productivity benefits of AI. In fact, the Australian Business Council last month said AI is the only biggest opportunity for a generation to increase productivity.
But what do we really know about how AI affects productivity?
What is productivity?
Simply put, productivity is the amount of amounts (products and services) that can be produced from a certain amount of input (such as labor and raw materials). That's important because higher productivity usually leads to a higher standard of living. Productivity growth accounts for 80% of Australia's income growth over the past 30 years.
Productivity can be considered individuals, organizations, or nationals.
Individual productivity is how efficiently you manage the time and resources to complete a task. How many emails can I reply to in an hour? Can I check for defects in one day?
Organizational productivity is how well an organization is achieving its goals. For example, how many top-quality research papers are produced in research organizations?
National productivity is the economic efficiency of the country, often measured as gross domestic product per hour. It is effectively a collection of other forms. However, it is notoriously difficult to track how changes in productivity for individuals or organizations change to national GDP per hour.
AI and personal productivity
Early studies examining the relationship between AI and individual productivity show mixed results.
The 2025 real-world study of AI and productivity involved 776 experienced product experts from the US multinational company Procter & Gamble. This study showed that randomly assigned individuals were assigned to use the AI performed and two-person team. A similar study in 2023 with 750 consultants from the Boston Consulting Group found that the task was 18% faster with the generation AI.
The 2023 paper reports on the early generation AI system of the Fortune 500 software company, which is used by 5,200 customer support agents. The system showed an increase of 14% in the number of problems resolved per hour. For less experienced agents, productivity increased by 35%.
However, AI does not always increase individual productivity.
A study of 2,500 experts found that the generator AI actually increases workload for 77% of workers. About 47% said they didn't know how to unlock the benefits of productivity. This study points to barriers such as the need to validate and/or modify AI output, the need for AI upskills, and the irrational expectations of what AI can do.
A recent CSIRO study examined the daily use of Microsoft 365 Copilot by 300 employees of government organizations. Self-report productivity for the majority was profitable, but not a significant minority (30%). Even workers who reported increased productivity expected a higher productivity advantage than offered.
AI and organizational productivity
If not impossible, it is difficult to attribut the changes in organizational productivity to the adoption of AI. Companies are sensitive to many social and organizational factors, one of which can be the reason for changes in productivity.
Nevertheless, the Organization for Economic Co-operation and Development (OECD) estimates the productivity benefits of traditional AI, namely machine learning benefits applied to industry-specific tasks, from zero to 11% at the organizational level.
The 2024 summary paper cites independent studies showing increased organizational productivity from AI in Germany, Italy and Taiwan.
In contrast, a 2022 analysis of 300,000 US companies found no significant correlations between AI adoption and productivity, but for other technologies such as robotics and cloud computing. Perhaps the explanation is that it is too difficult to solve the impact of AI, given that AI has not yet had an impact on many companies, or simply not applied alone.
Increased AI productivity can also be masked by the additional human labor required to train or operate AI systems. Take a walk through technology for Amazon for your shop.
It was released in 2018 and was intended to reduce labor as customer purchases were fully automated. However, it reportedly relied on hiring around 1,000 workers in India for quality control. Amazon labels these reports “false.”
More generally, think about the unknown number (but perhaps millions) you paid to label the data in your AI model.
AI and national productivity
National level photos are even more muddy.
Obviously, AI has not yet affected country productivity. It can be argued that technology development takes time to develop countries as companies need to know how to use technology and implement the necessary infrastructure and skills.
However, this is not guaranteed. For example, there is a consensus that the internet has led to increased productivity, but the impact of mobile phones and social media is more contested, and their impact is more evident in some industries (such as entertainment) than in other industries.
Productivity isn't just about making it faster
The general narrative about AI and productivity is that AI automates mundane tasks, makes things faster and more time for creative pursuits. But this is a simple view of how work happens.
Just because you can deal with your inbox more quickly doesn't mean you're spending the afternoon on the beach. The more emails you fire, the more replies you will receive and the never-ending cycle will continue.
It's not always better and faster. Sometimes we need to slow down to be more productive. That's when a great idea happens.
Imagine a world where AI is not just about speeding up tasks, but actively slowing us down and providing space to become more innovative and productive. It is a real, unexplored opportunity for AI.
This article will be republished from the conversation under a Creative Commons license. Please read the original article.
/Public release. This material of the Organization of Origin/Author is a point-in-time nature and may be edited for clarity, style and length. Mirage.news does not take any institutional position or aspect, and all views, positions and conclusions expressed here are the views of the authors alone.
