What We Are Finding


Reclaiming choice: AI, workforce displacement, and the community sector.

March 2026

Media and workforce response: ABC 7.30 / Diary of a CEO with Steven Bartlett

In the same week in March 2026, a former PwC consultant told the ABC that organisations not using AI will be overtaken, and an investigative journalist told a global podcast audience that workers are already losing bargaining power before AI can even do their jobs. Not For Humans examines what the 2026 AI workforce narrative means for community organisations, and why reclaiming the right to choose matters more than the choice itself.

ABC 7.30, AI, jobs and the workforce, 30 March 2026.

The Narrative


The language of inevitability

In late March 2026, the ABC’s 7.30 program aired a feature on AI and Australian workplaces. Among those interviewed was a former PwC consultant who had spent three years building AI agents for large corporations. His message was direct: organisations not integrating AI will be overtaken by those that do. Workers not using AI will fall behind those who are.

This is the dominant public narrative around AI and work in 2026. It is urgent, it is binary, and almost entirely framed around productivity and competitive advantage. What it does not ask:

  • Who is setting those terms
  • Who benefits from the urgency
  • What it closes off for organisations not positioned to adopt on demand

The same interview surfaced something more uncomfortable. The consultant described building a Microsoft Teams agent deliberately designed to impersonate a human colleague, updating a team without disclosing it was automated. He called it a gut punch morally. That detail did not make the headline. The adopt or die framing did.

The question is not whether AI will change work. It will. The question is who gets to decide how, at what pace, and with what protections for the people most affected.

The Research


When perception becomes reality

In the same week, investigative journalist Karen Hao appeared on the Diary of a CEO podcast drawing on her book Empire of AI, built from over 250 interviews with OpenAI employees and global AI researchers. One observation from that conversation cuts to the heart of what is happening in workplaces right now.

Hao argues that AI does not need to be capable of replacing a worker for displacement to occur. If a worker believes they can be replaced, and if the organisation believes it too, the bargaining dynamic shifts regardless of whether the technology can actually perform the work. Workers lose leverage. Organisations make cuts they might not otherwise make. Displacement precedes capability. The narrative does the work the technology has not yet done.

For community organisations, this dynamic is not abstract. It is arriving in the form of funder expectations, peer pressure from within the sector, and a creeping sense that organisations not visibly embracing AI are somehow behind. That pressure is real. But pressure is not the same as a governance decision. And a governance decision is what the moment requires.

This dynamic has a name in the Not For Humans framework: Technological Determinism. The assumption that technology follows a fixed trajectory and human adaptation is simply required. That framing serves the interests of those deploying the technology. It does not serve the interests of the communities depending on the people doing the work.

Displacement can precede capability. The narrative does the work the technology has not yet done. Community organisations need to be able to name this dynamic before they can respond to it.

The scale of the shift

59,000+

tech workers lost jobs in early 2026

AI explicitly cited as a driver in at least 1 in 5 cases. The roles disappearing follow a clear pattern: coordination, support, administration, and knowledge work.

IBTimes / RationalFX, March 2026

14%

drop in hiring of 22 to 25 year olds into AI-exposed roles

The pipeline of new workers developing skills in coordination and knowledge roles is already contracting. What this produces in three to five years is a workforce gap that cannot be filled from documentation alone.

Massenkoff and McCrory, Anthropic Institute, March 2026

What This Means Here


The community sector is not the tech sector

Research from the University of Sydney mapping Australian job listings against AI automation potential found that the roles most resilient against displacement share a common characteristic: they are built around physical presence, sustained human relationships, and embodied judgement developed over time.

Most resilient roles

  • Carers and community support workers
  • Allied health practitioners — physiotherapy, occupational therapy
  • Tradespeople and physically present roles
  • Roles centred on human connection and trust

Much of what community organisations do sits in that category. The relational, trust-based, advocacy work that defines the sector is not the same as the back-office administrative work most exposed to current AI capability. But community organisations carry both. The roles keeping organisations operational are doing exactly the kind of coordination and knowledge work AI tools are now moving into.

Roles most exposed in community organisations

  • Program coordinators
  • Case managers and intake officers
  • Grant writers and funding administrators
  • Communications and administrative staff

And here is where the community sector faces something the tech sector does not. The people in those roles are not just workers. They are often from the same communities the organisation serves. They carry relationships, local knowledge, and trust that cannot be documented or transferred to a system. When those roles are threatened, or when the people in them are made to feel expendable by a narrative that was never written with them in mind, something more than a job is at stake.

Questions worth asking in your organisation right now

  • → Who is making AI-related decisions in your organisation, and do they understand what is at stake for your workforce?
  • → Have the people whose roles are most affected had any meaningful say in how AI tools are being introduced?
  • → Is your organisation responding to genuine governance reasoning, or to the pressure of a narrative?
Reclaiming the right to choose is not about refusing AI. It is about insisting that community organisations, and the people who work in them, are participants in decisions that will reshape their work, not subjects of them.

What This Means Here


The community sector is not the tech sector

Research from the University of Sydney mapping Australian job listings against AI automation potential found that the roles most resilient against displacement share a common characteristic: they are built around physical presence, sustained human relationships, and embodied judgement developed over time.

Most resilient roles

  • Carers and community support workers
  • Allied health practitioners — physiotherapy, occupational therapy
  • Tradespeople and physically present roles
  • Roles centred on human connection and trust

Much of what community organisations do sits in that category. The relational, trust-based, advocacy work that defines the sector is not the same as the back-office administrative work most exposed to current AI capability. But community organisations carry both. The roles keeping organisations operational are doing exactly the kind of coordination and knowledge work AI tools are now moving into.

Roles most exposed in community organisations

  • Program coordinators
  • Case managers and intake officers
  • Grant writers and funding administrators
  • Communications and administrative staff

And here is where the community sector faces something the tech sector does not. The people in those roles are not just workers. They are often from the same communities the organisation serves. They carry relationships, local knowledge, and trust that cannot be documented or transferred to a system. When those roles are threatened, or when the people in them are made to feel expendable by a narrative that was never written with them in mind, something more than a job is at stake.

Questions worth asking in your organisation right now

  • → Who is making AI-related decisions in your organisation, and do they understand what is at stake for your workforce?
  • → Have the people whose roles are most affected had any meaningful say in how AI tools are being introduced?
  • → Is your organisation responding to genuine governance reasoning, or to the pressure of a narrative?
Reclaiming the right to choose is not about refusing AI. It is about insisting that community organisations, and the people who work in them, are participants in decisions that will reshape their work, not subjects of them.

Not For Humans

The for-purpose sector deserves governance tools built for its reality.

The AI workforce narrative is moving fast. The Not For Humans Toolkit gives community organisations a practical, self-guided way to build governance at the layer where it actually matters: human behaviour and decision-making. Before the pressure becomes the policy.

Not For Humans: Because AI’s greatest risk isn’t technical, it’s Human.

Not For Humans logo
Home » Australia’s National AI Plan