In Brief 

We are now at a point where artificial intelligence can be deployed to facilitate employers screening job candidates by scanning publicly available information without human intervention; tracking worker productivity with tracking devices and sensor cameras and automating redundancy selection processes by using pre-set data selection matrix, all without human input.  

It is therefore timely to consider the legal and ethical dimensions of a futuristic workplace where AI takes an increasingly prominent role – especially in light of the development of laws in jurisdictions such as Australia.  

What is happening in this space in Australia?

In February 2025, the Australian House of Representatives Standing Committee on Employment, Education and Training (Committee) tabled The Future of Work report (Report), following an inquiry into the digital transformation of workplaces. The Report makes twenty-one recommendations that focus on maximising the benefits of artificial intelligence (AI) and automated decision making (ADM) in the workplace, while attempting to address the risks and ensuring sufficient employee protections. 

The use of AI and ADM in the workplace requires close consideration, given the multitude of legal risks and regulatory issues relating to the laws of Australia – in particular at the federal level, such as the Privacy Act 1988 (Cth) (Privacy Act), the Fair Work Act 2009 (Cth) (Fair Work Act) and modern awards (which operate in addition to state and territory based laws).

AI and ADM in the workplace

Regulation of digital transformation in Australia continues to be an ongoing challenge across many industries, as the legal system plays catch-up on rapid technological advancement. The Committee‘s inquiry exposes gaps in Australia’s regulatory frameworks and workplace protections, at a time where there is a “very concerning and excessive” use of technology-enabled surveillance and data collection by employers.

Employers are increasingly using AI and ADM across all stages of the employment lifecycle, ranging from recruitment (screening and identifying candidates based on publicly available information) to training and performance management (tracking worker locations and productivity, detecting potential fraud and implementing security with digital identity verification, recording meetings amongst workers,  tailoring training to particular workers), to termination (automating determinations of redundancy). 

However, the use of AI and ADM is not without its legal risks, for both employers and workers. Key risks include:

• Ethics, including algorithmic bias and discrimination, privacy, and confidentiality: the output for AI and ADM can only be as good as the input, and where decisions are based on specific datasets, discrimination can become learned. The use of employee data by AI and ADM systems can also lead to misuse and violation of employee privacy. Similarly, AI has the potential to inadvertently breach confidentiality by mishandling sensitive information including of customers or suppliers;
• Transparency: if workers do not understand how decisions are being made, or feel they do not have control, they may lose trust in the employer.  Additionally, across Australia are a patchwork of laws dealing with workplace surveillance – some of which require notice, consent or other transparency measures to be put in place;
• Inaccuracies: relying on AI without human oversight can lead to inaccuracies and errors, and omit key issues, particularly in legal documents. It is difficult for buyers of AI systems to obtain information from the vendors on the reliability and accuracy of the system itself – and there may be other technology challenges once the system uses live information (which may differ from the training sets). Hallucinations by AI systems (that is, where the system generates incorrect, misleading, or nonsensical information, often presented as fact, due to flaws in its training data, biases, or processing errors) are not uncommon;¹
• Job displacement: as AI and ADM become increasingly efficient at automating tasks, there are concerns regarding workers being replaced, particularly in industries that rely on routine, repetitive, and/or manual labour.   

The Committee states, “employer decision making, which can be informed by technology, can have life-altering impacts on workers and their families, including termination of employment.”   

A key question will be precisely who is liable, and how, for discrimination and contraventions of workplace rights that occur as a result of AI and ADM.

Regulation of AI and ADM in Australia and elsewhere

Australia has no specific statute or regulations directly regulating AI. Instead, Australia relies on a patchwork of AI Ethics Principles and general laws contained in the Privacy Act and workplace surveillance legislation, which can differ greatly between states. There has been some enforcement of workplace surveillance laws in respect of the collection of  biometric information in the workplace.²

There is also movement towards reform and specific legislation at the federal level: 

• in September 2024, the Federal Government published the Voluntary AI Safety Standard (see Dentons’ article on the reforms) and is consulting on mandatory AI safety standards; and
• in November 2024, amendments to the Privacy Act were introduced which specifically address ADM using personal information that would have a significant impact on the individual. These amendments will take effect from 10 December 2026 (see Dentons’ article on the amendments).

Comparatively, the leading comprehensive regulatory framework is the AI regulation in the European Union (EU) which has implemented laws governing the development and use of AI (see the EU’s Artificial Intelligence Act, approved 21 May 2024 – these laws are discussed with other global regulatory developments on the Dentons AI Global Solutions Hub.

Report recommendations

In the face of rising AI and ADM use, the Report makes key recommendations, which include that the Australian Government:

1. classify AI systems used for employment related purposes as high-risk and adopt the proposed mandatory guardrails for high-risk AI;
2. review the Fair Work Act to ensure decision marking using AI and ADM is covered, and that employers remain liable for these decisions;
3. review the National Employment Standards and modern awards to respond to the adverse effects of significant job redesign caused by emerging technologies;
4. review the Privacy Act and the Fair Work Act  to protect workers’ data and privacy, including banning high-risk uses of worker data, and empowering the Fair Work Commission to manage the complaints relating to breaches of workers’ privacy obligations;
5. work with states and territories to develop greater consistency and better protections against excessive and unreasonable surveillance in the workplace;
6. amend the Fair Work Act to improve transparency, including by requiring all organisations that use AI or ADM systems to disclose this fact to existing and prospective workers and customers.

What this means for employers in Australia

While it remains to be seen how many of the recommendations the Australian Government might adopt, employers in Australia can prepare their workplaces for greater AI and ADM regulation by:

• implementing an AI and ADM policy that addresses both how the employer uses AI in the workplace, and the obligations and expectations for employee use of AI and ADM;
• reviewing AI and ADM processes, including surveillance, to ensure that worker data and privacy is adequately protected and there is legal compliance with existing regulation;
• consulting with the workplace about the introduction or use of AA and ADM;
• ensuring that any major decisions relating to the employment relationship, such as termination, have human oversight to mitigate the risk of errors and inaccuracies, which may lead to claims and liabilities for employers;
• keeping a close watch on the legislative developments.

Please contact Paul O’Halloran, Robyn Chatwood or Michael Park or your usual contact in the Dentons Employment & Safety Team or Technology Team for further guidance.

1. See, for example, the hallucination index introduced by Galileo
2. See, for example, Jeremy Lee v Superior Wood [2019] FWCFB 2946 which dealt with use of fingerprint scanning in a workplace.