Australia released a draft workplace AI adoption guide for public consultation on 12 November 2026, setting out a voluntary approach to managing technological change with workers. For university employers and research offices, the draft brings staff participation, job design, training and accountability into the discussion alongside the choice of AI tools.
A workplace guide opens for public comment
Employment and Workplace Relations Minister Amanda Rishworth announced the draft Best Practice Guide on the Adoption of AI in the Workplace following roundtables and forums involving unions, employer representatives, academics, small businesses and large employers. The government presents it as an approach to sharing AI’s benefits between organisations and their workforces.
There are two relevant opening dates. Rishworth’s release and the Department of Employment and Workplace Relations’ public announcement are dated 12 November. The consultation hub timeline records an opening date of 11 November. Both point to the same consultation, which closes at 11:59 pm AEDT on 27 January 2027.
Voluntary guidance alongside existing obligations
The 45-page draft explicitly says it is voluntary, does not create or expand legal obligations, and is not a code of practice under work health and safety laws. It describes practices to consider alongside existing legal requirements, including workplace relations, privacy, intellectual property, cyber security, anti-discrimination and workplace safety.
It also distinguishes its purpose from the National AI Centre’s existing Guidance for AI Adoption. That framework contains six essential practices for responsible adoption across the economy. This workplace draft has eight principles, with a specific emphasis on employee engagement, changing roles and workforce transition. The two documents are intended to complement one another.
Its scope extends beyond generative chatbots: the draft discusses AI embedded in human resources platforms, rostering, productivity software and other workplace systems. It recognises that systems which assist or automate tasks can both change the responsibilities within a job.
Eight principles across planning, capability and oversight
The draft organises its principles into three parts. The structure links the initial decision to introduce AI with the way work subsequently changes and the oversight needed after deployment.
| Theme | Workplace principles | Focus in the draft |
|---|---|---|
| Plan | Capture the opportunity; identify use cases and assess readiness; plan cooperatively and engage workers. | Set goals, identify actual workplace problems and involve staff before significant decisions. |
| Enable | Design work and jobs; build capability; manage workforce transformation. | Consider task changes, skills and support for people whose roles evolve. |
| Embed | Establish AI governance; proactively manage risks and work health and safety. | Assign accountability, monitor systems and respond to workplace impacts. |
Under planning, the draft asks employers to draw on workers’ knowledge of bottlenecks and daily operations. Readiness includes digital infrastructure and data security, but also whether managers have time, resources and authority to engage staff and act on their feedback. Worker participation is described as an ongoing process with opportunities to influence decisions.
The capability section goes beyond learning to operate a tool. It includes critical evaluation of AI outputs, cyber security, organisational change, role-specific skills, technical implementation and governance, and human judgement. It also says capability needs will change as jobs, systems and organisational priorities evolve.
Governance continues after deployment
For AI governance, the draft favours integration with existing risk, legal and compliance structures. Humans remain accountable, with responsibilities and approval pathways proportionate to each system’s use and potential impacts. Its recommendations include reporting incidents and near misses, reviewing workload and job satisfaction, and using worker feedback to reassess risks.
The risk section identifies intrusive monitoring, unrealistic performance standards and work intensification among potential workplace hazards. It also addresses bias, errors and data quality. These provisions make the effects on people part of continuing oversight, alongside whether a system performs its intended function.
The government’s argument and initial coverage
In her 12 November AFR AI Forum speech, Rishworth argued that buying technology alone does not secure its benefits: workers need confidence in its use and a clear understanding of what changes mean for them. She also emphasised leaders’ ability to manage organisational change and continuing consultation about risks.
The National AI Centre reinforced that distinction on 13 November. Deputy director Brett Szmajda described the new draft as a closer examination of how businesses and workers plan change, cooperate and build skills alongside the existing responsible-adoption guidance.
Independent coverage identified the same development. ABC’s The World Today described a voluntary guide following months of consultation. The Mandarin’s Melissa Coade reported the public consultation and eight principles on 12 November.
What it means for universities and research offices
For universities, the practical implication is a broader workplace discussion when adopting AI. As an editorial application of the draft’s general provisions, a research office considering changes to administrative workflows would examine staff input, task allocation, training and escalation routes alongside technical capability. The draft offers a basis for those discussions; it sets no university-specific implementation deadline.
That perspective connects an institutional governance framework to the people doing the work. For research managers, the relevant questions include who approves a use case, who reviews its effects on workload, and whether staff can influence adjustments after deployment. These follow directly from the draft’s attention to accountability and continuing worker engagement.
It also adds a workplace dimension alongside AI-assisted drafting practices for researchers. Checking outputs is one capability; managing a changed job or team workflow is another. The draft’s separate capability domains give institutions a way to discuss those different training needs.
What happens next
The department’s invitation explicitly includes academics, alongside businesses, workers, unions, governments, not-for-profit organisations and the public. Feedback can be submitted through the consultation hub’s online form until 27 January 2027. The hub says individual submissions may be published. The immediate next step is consultation on the draft; the notice does not announce a final publication date.
Sources
- Ministers’ Media Centre, Draft best practice guide for AI adoption in the workplace released, 12 November 2026. Checked 2026-11-15.
- Department of Employment and Workplace Relations, draft guide consultation hub, timeline opened 11 November 2026. Checked 2026-11-15.
- Department of Employment and Workplace Relations, Best practice guide on the adoption of AI in the workplace, draft for consultation, November 2026, pp. 4–8, 17–21, 30–31 and 39–45. CC BY 4.0. Checked 2026-11-15.
- Department of Employment and Workplace Relations, consultation announcement, 12 November 2026. Checked 2026-11-15.
- Ministers’ Media Centre, Amanda Rishworth, AFR AI Forum speech, 12 November 2026. Checked 2026-11-15.
- National AI Centre, Brett Szmajda, Have your say on draft guidance for AI adoption in workplaces, 13 November 2026. Checked 2026-11-15.
- ABC, The World Today, Draft guide released for AI use, 12 November 2026. Checked 2026-11-15.
- The Mandarin, Melissa Coade, AI at work needs collaborative workplace transformation, government says, 12 November 2026. Checked 2026-11-15.
