Is the AI Productivity Argument Actually Over? An Australian SMB Reality Check
Joe Procopio declared it in Inc.com on 10 July: the AI productivity argument is over. His case is worth taking seriously. The aggregate economic data hasn’t shifted. Enterprise ROI on custom AI deployments is patchy at best. The 10x, 100x, 1000x promises from 2023 haven’t shown up in the GDP numbers. If your measure of AI success is “did it transform business output at scale,” then yes, the argument is hard to win right now. That’s a fair read of the US enterprise picture.
But here’s what that picture is actually showing: large companies spending big on custom AI builds, bespoke integrations, and internal AI teams chasing change projects. That is not the story for Australian SMBs. Not even close.
The Research Base Doesn’t Match
The studies Procopio draws on, and the broader US commentary around AI ROI, are almost entirely rooted in enterprise deployments. McKinsey surveys of companies with 10,000+ employees. Goldman Sachs productivity analyses of knowledge-worker roles in large organisations. MIT research on professional services firms. None of that maps onto a three-person accounting firm in Geelong using ChatGPT to write client emails, or a tradie in Brisbane generating quotes with an AI tool bolted onto their CRM.
Australian SMBs aren’t running change projects. They’re using off-the-shelf tools, today, to do specific things faster. The research base arguing AI productivity has failed simply isn’t studying them.
What the Australian Data Shows
Between 37% and 68% of Australian SMEs are now using AI tools, depending on which survey you read, MYOB’s Bi-Annual Business Monitor, CSIRO’s tracking work, and the National AI Centre’s adoption data all land somewhere in that range. The variation mostly reflects how “AI tool” gets defined. Whether you take the conservative number or the generous one, the trend is the same: adoption is up, fast, and it’s not slowing.
More telling is what SMBs are reporting. In surveys of Australian AI users, 79% report productivity improvements. Not “transforming the business”, getting specific things done faster. Emails drafted in half the time. Quotes turned around the same day instead of three days later. Social posts that used to take two hours now take twenty minutes.
That’s the story Inc.com missed: the productivity argument isn’t about enterprise change. It’s about getting two hours back on a Tuesday.
Where Inc Is Right
To be straight about it: Procopio has a point in specific contexts. If an Australian SMB owner is being pitched a $50,000 custom AI build, a bespoke chatbot, a custom-trained model, a proprietary integration project, healthy scepticism is warranted. Enterprise-grade deployment risk applies regardless of business size. Custom projects take longer than promised, cost more than quoted, and often deliver less than the demo suggested.
The argument Procopio is really making is: don’t believe the hype when someone else controls the scope, the timeline, and the invoice. On that, he’s right.
The mistake is applying that caution to everything. A $30-a-month subscription to an AI writing tool is not a change project. It’s a tool. The ROI question is simple: does it save enough time in the first month to pay for itself? Usually the answer is yes after day three.
Two Questions That Cut Through the Noise
Before believing any AI productivity claim, from a vendor, a podcast, a consultant, or an Inc.com column, two questions are worth asking.
First: is this a tool I can use in the next 20 minutes, or a project? Tools have immediate feedback loops. If it takes a week to set up, a specialist to configure, and a meeting to evaluate, you’re in project territory. Project risk is real. Tool risk is low.
Second: does the time saving pay for itself in month one? Not year one. Month one. If the answer isn’t obvious within thirty days, the value probably isn’t there, or you haven’t found the right use case yet.
Most of the AI tools Australian SMBs are actually using pass both tests easily. The ones that don’t probably shouldn’t be marketed as productivity tools in the first place.
The Argument Hasn’t Started Yet
The AI productivity argument isn’t over. It’s just been held in the wrong court.
US enterprise AI deployments failing to move the macro needle is a real phenomenon worth understanding. But it’s a different conversation to whether a Melbourne bookkeeper who started using AI to write client update emails in January is getting value from it. She is. Her clients get faster responses. She leaves the office earlier on Fridays. That’s not a headline anyone at Goldman Sachs is going to write a note about.
The right question for Australian SMB owners isn’t “has AI lived up to its promises?” That’s a question for the enterprise market that made the promises. The right question is: which tasks in my day could I hand off to a tool that costs less than a coffee subscription?
That conversation has barely started. And it’s the only one that matters here.
📖 More deep dives: News Deep Dives
📊 Compare AI tools side by side | 💼 Free resources & AI prompt packs
📬 The SmallBizAI Brief
One practical AI tip for Australian small business. Every Tuesday. Free.
Join business owners getting smarter about AI every week.
More hot takes: Hot Takes →
