(Cth) Include ADM as a Judicially Reviewable Decision

Download the 1-Page Brief Here - (Cth) Include ADM as a Judicially Reviewable Decision.pdf

Download PDF • 187KB

Author: Nicole Brideson & Amber Nguyen | Publish date: 3/3/2026

• P: It is unclear whether a decision made wholly or partly by an automated system on behalf of a Commonwealth authority constitutes a judicially reviewable decision. 

• S: The Attorney-General should amend Section 3 of the Administrative Decisions (Judicial Review) Act 1977 (Cth) to clarify that a ‘decision’ includes one wholly or partly made by an automated system.

Problem Identification: 

Section 3 of the Administrative Decisions (Judicial Review) Act 1977 (Cth) (the Act) states that a Commonwealth authority’s administrative ‘decision’ can be subject to judicial review. Common Law has interpreted a ‘decision’ under the Act to involve a ‘mental process’ of reasoning to reach a conclusion. However, the Act does not specify whether decisions made using automated systems are included.

According to Prof. Huggins, from the QUT Faculty of Law, ‘[t]here is thus a mismatch between ADM [Automated Decision-Making] … and the legal meaning of a decision.’ She explained that ‘… fully automated discretionary decisions may not be reviewable under [the Act]’. This may result in individuals affected by a Commonwealth authority’s decision being at risk of having no legal redress.

Context: 

The Act applies to administrative decisions authorised under all Commonwealth Acts (e.g.  some Child Support, Renewable Energy and Aged Care Acts) unless explicitly excluded, and also applies to some state Acts outlined in Schedule 3. Remedies available under the Act enable the court, among other things, to set aside a decision or send it back to the original decision-maker. 

ADM refers to a system which, ‘by its very nature, requires little or no human input after the initial coding decisions have been made’.

According to Senior Counsel Hogan-Doran, the judicial review landscape has shifted from ‘a world in which governmental activity was populated by human policy makers’ to one of ‘automated decision-making, driven by algorithms’. For example, a report by the Office of the Australian Information Commissioner (OAIC) found that of 23 government agency websites reviewed, ‘17% of agencies disclosed the use of ADM in decision-making’, whilst ‘9% were identified as “likely to be using ADM”’ but had not disclosed use.

In 2018, the case of Pintarich v Deputy Commissioner of Taxation set a precedent concerning the use of ADM. The Federal Court found that a letter issued by an automated system from the ATO, bearing the Deputy Commissioner’s signature, was not a decision that could be reviewed, as there was no ‘mental process’ behind the letter. Justice Kerr, in dissent from the majority, highlighted that the decision ‘[would be] productive of “administrative uncertainty and confusion”.’ 

Arguments:

The Human Technology Institute noted that the Act could prevent fairness and certainty for individuals in reviewing automated administrative decisions. They explained this is because individuals may be unable to ‘rely on communications from government’. For example, according to Prof. Huggins, individuals could receive automated correspondence that grants a benefit, or requires payment of a debt, only to be told the automated notice was legally inoperative. The Australian Human Rights Commission similarly cautioned that it would be unfair if a person had acted in reliance on these inoperative notices. Dr Ng, Associate Prof. at Monash University Faculty of Law, and Dr O’Sullivan, Associate Prof. at Deakin Law School, considered that without reform, ‘individuals are precluded from a major avenue of challenge.’

Dr Ng warned that there may be ‘a perverse incentive for departments and agencies to automate so as to avoid judicial review.’

Justice Kerr posited, ‘[i]t would undermine fundamental principles of administrative law if a decision could [be renounced] … simply by asserting there was a distinction between their mental processes and the[ir] expression’. Following the Robodebt Royal Commission, Dr Ng and Dr O’Sullivan recognised a heightened need for ‘legality and accountability’ in automated government processes. 

Dr Ng and Dr O’Sullivan stated that the law is falling behind technological developments. Justice Kerr argued that ‘what constitutes a decision cannot be static; it must comprehend that technology has altered how decisions are in fact made and that … decision making, can occur independently of human mental input.’ Further, Prof. Huggins identified that ‘mismatches between the law and technology impede opportunities for meaningful legal accountability for errors in [ADM].’

Advice/Solution Identification:

Dr Ng has called for the definition of a ‘decision’ in the Act to be amended to clarify that it includes decisions made wholly or partly by an automated system. She reasoned this would provide ‘a one-step solution’ without requiring each piece of legislation to be amended. 

Further, UNSW Allens Hub et al. argued in their Submission to the Robodebt Royal Commission, ‘[i]f a government department outsourced some of its functions to a supplier, the department would remain accountable for those functions. The same should be true of a department which outsources its functions to a machine.’

Precedent:

There is domestic and international precedent for recognising ADM as reviewable. Domestically, legislation such as the Paid Parental Leave Act 2010 (Cth) and the Veterans’ Entitlements Act 1986 (Cth) has already clarified that ADM decisions are those of the authorised officer, and therefore reviewable. Internationally, the EU recognised a right to contest ADM decisions. A 2023 EU case held that individuals could challenge decisions based on automated processes.

---

Public Support: 

1. Dr Yee-Fui Ng

2. Human Technology Institute

Broad Support:

1. Australian Human Rights Commission - They have broadly called for ‘relevant legislation’ to be changed to clarify that ‘for the avoidance of doubt, the term ‘decision’ includes decisions made using automation and other forms of artificial intelligence’. But, they have not explicitly specified that this would amend the Administrative Decisions (Judicial Review) Act 1977 (Cth).

   
   

This list reflects publicly stated positions and should not necessarily be taken as endorsement of this specific brief.

News Coverage:

• ABC News - “Tax office computer says yes, Federal Court says no”. This article covered an analysis of the Pintarich precedent, and predicted its roll-on effects. By: Nassim Khadem | 5 October 2018 - Read the article here.

• ABC News - “Automatic systems unlawfully cancelled 964 jobseekers' payments in two years, watchdog finds”. The article reported on the cancellation of Job Seeker Payments through the use of the Targeted Compliance Framework, an automated system intended to ensure mutual compliance. By: Maani Truu | 6 August 2025 - Read the article here.

• The Law Report, ABC Radio National - “AI and automated decision making in government”. A podcast that discussed the potential checks and balances that could be developed to facilitate the use of AI in government operations. By: Damien Carrick | 8 July 2025 - Listen to the episode here.

• The Conversation - “Most Australian government agencies aren’t transparent about how they use AI”. The article analysed the transparency reporting adopted by government agencies regarding AI and ADM. It included comparisons to other international jurisdictions. By: José-Miguel Bello y Villarino, Alexandra Sinclair and Kimberlee Weatherall | 27 October 2025 - Read the article here.

Where to go to learn more: 

1. (2025) Dr Yee-Fui Ng - This submission to the Attorney-General’s Department’s consultation on the use of ADM by Government Agencies addressed the uncertainty created by the Pintarich precedent and recommended amendments to legislation, including the Administrative Decisions (Judicial Review) Act 1977 (Cth), to clarify that a ‘decision’ included one generated by an ADM to address that uncertainty. Find their submission here. 

2. (2025) Human Technology Institute - This submission to the Attorney-General’s Department’s consultation on the use of ADM by Government Agencies called for a rectification of the uncertainty generated by the Pintarich precedent by implementing a change to the Administrative Decisions (Judicial Review) Act 1977 (Cth) to clarify that a ‘decision’ is one made wholly or partly by an automated system. Read their submission here.

3. (2025) Australian Human Rights Commission - This submission to the Attorney-General’s Department’s consultation on the use of ADM by Government Agencies explored the ways that ADM could be regulated to ensure compliance with Human Rights. It proposed amendments to legislation to clarify that a ‘decision’ included one generated by ADM. Read their submission here.  

4. (2025) Automated Decision Making and Deportation: Legal Concerns and Regulation - George Yijun Tian, Tim McFarland & Sanzhuan Guo - An academic exploration which analysed the use of ADM systems when making immigration decisions and determinations. It suggested regulatory responses to address the various risks/issues associated with ADM implementation, and it proposed a ‘simple schema’ to group the risks and improve how these systems are regulated. Read it here.

5. (2020) Revitalising Public Law in a Technological Era: Rights, Transparency and Administrative Justice - Yee-Fui Ng, Maria O’Sullivan, Moira Paterson & Normann Witzleb - An academic article which addressed developing challenges faced by administrative law and how ADM could be utilised and regulated to protect the integrity of public law and provide fairness to individuals affected by administrative decisions. Read it here.

6. (2019) Deliberation and Automation - When is a Decision a “Decision”? - Yee-Fui Ng and Maria O’Sullivan - A seminal journal article which unpacked the Pintarich case and analysed ADM’s impact on the administrative landscape post Robodebt. It looked into the uncertainty of what made an administrative decision a reviewable one under the Act. Find it here.

7. Administrative Decisions (Judicial Review) Act 1977 (Cth) - Read the Act here.

Human Perspective: 

Sam operates a wind farm and applies for accreditation so he can contribute to Australia’s renewable energy targets. Under the Renewable Energy (Electricity) Act 2000 (Cth), the Regulator must grant accreditation if all eligibility requirements are met. Sam believes that he satisfies these requirements and, relying on approval, secures a substantial loan against his family home to fund the project. Shortly after applying, Sam receives a notice, signed by the Regulator, refusing accreditation. The refusal prevents him from operating the wind farm and places him at risk of defaulting on his loan. Seeking accountability, Sam applies for review under the Act responsible for reviewing these decisions. Only later does Sam learn that the refusal was generated entirely by an automated system. As the Act responsible for reviewing these decisions does not clearly recognise automated outcomes as ‘decisions’, the refusal may be found not to be reviewable at all, despite its serious financial consequences and official appearance. Sam is left without a clear avenue to challenge a government action that has directly jeopardised his livelihood and home.   

To protect the anonymity of those involved, this is a fictionalised account drawn from an amalgamation of real-life stories, experiences and testimonials gathered during the research process for this brief. Any resemblance to actual individuals is purely coincidental.

Conflict of interest/acknowledgment statement: 

N/A

Support 

If your organisation would like to add your support to this paper or suggest amendments, please email Info@foreaustralia.com. 

Disclaimers

Please review all FORE disclaimers here.

Reference list: 

Administrative Decisions (Judicial Review) Act 1977 (Cth). https://www.legislation.gov.au/C2004A01697/2014-12-16/text

Attorney General’s Department. (2024). Use of automated decision-making by government: Consultation paper. Australian Government. https://consultations.ag.gov.au/integrity/adm/user_uploads/consultation-paper-use-of-automated-decision-making-by-government.pdf 

Boardman, R., & Lölfing, N. (2023, December 19). Key takeaways from the CJEU’s recent automated decision-making rulings. International Association of Privacy Professionals. https://iapp.org/news/a/key-takeaways-from-the-cjeus-recent-automated-decision-making-rulings

Hogan-Doran, D. (2021). Computer says “no”: Automation, algorithms and artificial intelligence in government decision-making. In Handbook for judicial officers (pp. 1113-1143). Judicial Commission of New South Wales. https://jirs.judcom.nsw.gov.au/public/assets/benchbooks/judicial_officers/ 

Huggins, A. (2018, November 14). Automated processes and administrative law: The case of Pintarich. Australian Public Law. https://www.auspublaw.org/blog/2018/11/the-case-of-pintarich

Huggins, A. (2021). Addressing disconnection: Automated decision-making, administrative law and regulatory reform. University of New South Wales Law Journal 44(3), 1048-1077. https://www.austlii.edu.au/au/journals/UNSWLawJl/2021/37.html 

Huggins, A. (2025, November 24). Regulating automated government decision-making: An Australian perspective - Combatting the Code book forum. Australian Public Law. https://www.auspublaw.org/blog/2025/11/regulating-automated-government-decision-making-an-australian-perspective-combatting-the-code-book-forum

Khadem, N. (2018, October 5). Tax office computer says yes, Federal Court says no. ABC News. https://www.abc.net.au/news/2018-10-05/tax-office-computer-says-yes-federal-court-says-no/10341548 

New South Wales Ombudsman. (2024). A map of automated decision-making in the NSW public sector: A special report under section 31 of the Ombudsman Act 1974. https://cmsassets.ombo.nsw.gov.au/assets/Reports/Introduction-and-Brief-Observations.pdf 

Ng, Y.-F., & O’Sullivan, M. (2019). Deliberation and automation: When is a decision a “decision”? Australian Journal of Administrative Law, 26(1), 21–34. https://search.informit.org/doi/10.3316/agispt.20190404008510 

Ng, Y.-F. (2021, August 20). The rise of automated decision-making in the administrative state: Are Kerr’s institutions still ‘fit for purpose’? Australian Public Law. https://www.auspublaw.org/blog/2021/08/the-rise-of-automated-decision-making-in-the-administrative-state-are-kerrs-institutions-still-fit-for-purpose

Ng, Y.-F. (2021). Towards constitutional consonance: Institutional adaptation and the administrative state. Melbourne University Law Review, 44(3), 889-927. https://www.austlii.edu.au/cgi-bin/viewdoc/au/journals/MelbULawRw/2021/4.html 

Ng, Y.-F. (2025, November 28). Combatting the Code book forum — Author response. https://www.auspublaw.org/blog/2025/11/combatting-the-code-book-forum-author-response

Ng, Y.-F. (2025). Submission to the Automated Government Decision-Making Consultation. Retrieved January 10, 2026, from https://consultations.ag.gov.au/integrity/adm/consultation/view_respondent?uuId=797241227

Perry, M., & Campbell, S. (2021). AI and automated decision-making: Are you just another number? [Speech transcript]. Federal Court of Australia. https://www.fedcourt.gov.au/digital-law-library/judges-speeches/justice-perry/perry-j-20211021

Pintarich v Deputy Commissioner of Taxation [2018] FCAFC 79. https://www.austlii.edu.au/cgi-bin/viewdoc/au/cases/cth/FCAFC/2018/79.html?context=1;query=pintarich%20v%20deputy%20commissioner%20of%20taxation%20;mask_path=

Regulation 2016/679 (General Data Protection Regulation). Article 22: Automated individual decision-making, including profiling. European Union. https://gdpr-info.eu/art-22-gdpr/

Senate Select Committee on Adopting Artificial Intelligence. (n.d.). Chapter 5 — Automated decision-making [Final report]. Parliament of Australia. Retrieved January 10, 2026, from https://www.aph.gov.au/Parliamentary_Business/Committees/Senate/Adopting_Artificial_Intelligence_AI/AdoptingAI/Report/Chapter_5_-_Automated_decision-making