An Australian news channel aired a roundtable where government ministers sat alongside executives from the world's largest AI companies to discuss the future of employment. There's something familiar about the theater of it. People in ties solemnly debating what to do about a problem that has, in large part, already happened. While the microphones captured promises of regulatory frameworks and orderly transition, the call centers of Sydney and Melbourne had spent months cutting staff. They replaced human operators with conversational agents that answer in seconds and never ask for a holiday bonus.

That gap between public debate and fait accompli is the real story here. Automating service work alters the relationship between human effort and economic value because it strips workers of their collective bargaining power. Their ability to say \"this doesn't work without me\" gets transferred to software that needs no union and no rest. This is the basic fact worth pinning down before going further. It's not just about losing jobs. It's about losing the leverage that a job gave ordinary people against those who control capital.

Australia became the stage for this debate for concrete reasons. Its service economy relies heavily on customer support and on higher education as an export. The World Economic Forum flags both sectors as among the most exposed to task reconfiguration by generative AI. The country also has a track record of attempting to regulate disruptive technologies with some ambition. The results remain debatable.

What caught attention wasn't so much the content of the debate as who was sitting there. Global AI leaders discussing the future of work with ministers is an image we've already seen play out elsewhere. In \"UBI: When Those Who Cause Unemployment Sell the Solution,\" I documented how Sam Altman and Elon Musk pushed universal basic income while their own companies executed tens of thousands of layoffs and poured hundreds of billions of dollars into AI infrastructure. The pattern repeats. Whoever designs the automation also gets a seat at the table designing the cushion that's supposedly meant to soften it. That's not necessarily bad faith. It's a conflict of interest that rarely gets mentioned in the press releases.

This brings up a distinction I've been tracking across different contexts for a while now: solving a problem versus merely managing it. A minister sitting down with an AI company executive to discuss national standards is, in theory, looking for a solution. But if the party that benefits most from frictionless automation ends up drafting the standard, nothing gets solved. It just formalizes the status quo in the language of public policy. The Generosity in the Doorway makes a point that applies directly here. The debate over which policy is best can run forever without ever touching the question that actually decides the outcome: who has the power to implement it, and who has the power to refuse.

Game theory applied to institutional design helps clarify this. When two actors negotiate under rules where one controls the technology and the other controls the regulation, the outcome depends less on who wins the argument at the table and more on the incentives each has off camera. A minister needs to show he's doing something before the next election. An AI executive needs to show investors that adoption won't trigger a regulatory backlash that slows growth. Neither one has, as a priority, solving the problem of the call center worker who has already lost their job. That worker isn't at the table. They never are.

Are there examples of communities that managed shared resources without a single group hoarding all the decision-making power? Elinor Ostrom documented communities that designed their own rules for using water, forests, and fisheries. These systems worked for generations. Simple rules. Lasting results. The people who used the resource took part in designing the rules. In the case of AI and employment, the resource is controlled by a handful of companies. The displaced user isn't even invited to the table where their future gets decided.

For someone working today in a call center in Brisbane or Perth, this has direct consequences. The ongoing training companies offer as an alternative is often cosmetic if it doesn't come with real bargaining power. It doesn't change the balance of forces if that same software eventually makes the operator unnecessary anyway. Training people for a job market that doesn't have enough positions doesn't solve the structural problem. It only delays it and individualizes it, as if the lack of jobs were a skills problem rather than a matter of economic architecture.

I don't have a tidy answer for what should happen at that Australian roundtable. It would be dishonest to pretend otherwise. The regulatory frameworks being discussed in Canberra will likely reflect the interests of those who already hold the pen, unless some organized force emerges capable of contesting that authorship. That hasn't happened yet in any country.

The table should include the call center operators, the teachers whose evaluations are being redefined, and the people who will actually live with the consequences. The conversation would sound different, and it would make more than a few of the current attendees uncomfortable.

How would the debate change if the people directly affected had a voice and a vote in it?

Sources:

1. World Economic Forum, Future of Jobs Report (recent editions on the transformation of workplace skills)

2. Elinor Ostrom, Governing the Commons (theoretical framework on shared resource management)

3. Yves Laurent, The Generosity in the Doorway (Chapter 14 — The Question Beneath the Question)

4. Previous article by the author: \"UBI: When Those Who Cause Unemployment Sell the Solution\" (2026-06-24)

5. Previous article by the author: \"Jóvenes Construyendo el Futuro: The Scholarship That Creates No Jobs\" (2026-08-13)