Abstract
We are delighted to introduce the 2025 Annual Review of Critical Developments in Industrial Relations. As previously, this year's Review is focused on a theme selected with assistance from the Journal's Industry/Practitioner Advisory Committee, whose members identified the need for a dedicated review of developments relating to Artificial Intelligence (AI) and its impact on work, working people and industrial relations. As usual we include a practitioner contribution and we thank Osmond Chiu and Katie Higgins from the Community and Public Sector Union, whose paper ‘AI in the Australian Public Service: What federal public servants really think about it’, provides an overview of a union survey of nearly 2000 Australian Public Service (APS) employees. 1 Conducted from late 2024, the survey found that AI use across the APS is generally viewed positively. However, the data also raised concerns about a lack of consultation and knowledge about AI governance arrangements, inadequate training for staff, and the use of AI in recruitment. Another key issue raised centred on fears about its impact on public trust in government. Trust is also addressed by Andreas Cebulla. In this paper entitled, ‘AI and workplace relations: a work health and safety (WHS) framework for managing relational risks in workplaces’, Cebulla questions the assumed productivity gains from AI by focusing on relational risks affecting workplace dynamics and identifying gaps in Australia's policy response particularly in regard to the integration of AI-related risks into WHS regulations. An important aspect of this paper is a proposed framework for managing relational risks grounded in job crafting, participatory oversight and expanded WHS definitions, which aims to position the worker not as a passive recipient of AI impacts but as a co-designer of workplace transformation.
Cebulla is not alone in considering AI's impact on relations among AI users in the workplace. An article in the Harvard Business Review in late September 2025 entitled: ‘AI-generated “workslop” is destroying productivity’, outlined results of a study that found that many ‘organizations see no measurable return on their investment from fully AI-led processes.’ According to this report's authors, one possible reason is that ‘[e]mployees are using AI tools to create low-effort, passable looking work that ends up creating more work for their coworkers’. In their view, such ‘AI generated work content that masquerades as good work but lacks the substance to meaningfully advance a given task’, offloads cognitive labour onto coworkers, costing ‘nearly two hours of rework per instance’ and creating ‘downstream productivity, trust, and collaboration issues' (Niederhoffer et al., 2025). The concerns raised in this report were subsequently picked up by numerous other publications, such as the Huffington Post (Glover, 2025) and the Guardian (Marks, 2025), with the latter reporting on a KPMG study that echoed trust concerns as well as widespread support for regulation (Gillespie et al., 2025: 5).
Questions of AI regulation have been prominent on the agenda in Australia, particularly following developments in the European Union (EU). Over the past 2 years, the Australian Government's Productivity Commission (2024a, 2024b, 2024c) has published a number of papers on AI opportunities, uptake, productivity, the role of government, challenges of regulating AI, and data policy. Most recently, the Commission ‘opposed the introduction of tough laws to control AI being considered by the government, warning its plan for “mandatory guardrails” should be paused until gaps in the law are properly identified’ (Evans, 2025). This position aligns with the Business Council of Australia's Accelerating Australia's AI Agenda, released in June 2025, which encompasses a strong focus on a ‘productivity imperative’ and focuses on numerous AI challenges. Among these, it argues that ‘in the age of AI’ regulation ‘is a tightrope walk’. While recognising that rules are needed ‘that build public trust and protect citizens from genuine harm, such as deepfakes manipulating opinions, biased algorithms perpetuating inequality, or breaches of data privacy’, it stressed that ‘[w]e must avoid strangling the very innovation we seek to foster. Overly burdensome, ambiguous, or premature regulations throttle experimentation, deter investment, and put Australian businesses at a disadvantage.’ In this regard, it specifically highlights ‘the EU's regulatory burden’ hindering ‘the EU's competitiveness and performance’ (Business Council of Australia, 2025a: 16, 26; Also see Business Council of Australia, 2025b).
Several papers in this Annual Review directly reflect on such challenges. One, entitled ‘Rethinking Industrial Relations: Policy Ecologies, Cultural Work, and Artificial Intelligence’, by Kate MacNeill, Amanda Coles and Gareth Fletcher, examines AI in relation to the creative industries and maps the fragmented governance of AI vis-à-vis creative labour across industrial relations, copyright law, and cultural policy, arguing that a reimagining of industrial relations is required to ensure that such labour is adequately protected in an economy increasingly driven by AI. Another by Penny Williams, entitled, ‘More than AI: The Platformisation of work’ refers to the use of app-based and AI-driven platforms to automate human resources practices in standard work arrangements in Australia and argues that technologies enabling Algorithmic Management have already been eroding conditions of work, worker rights and labour relations with limited regulatory or policy intervention. Finally, our Controversy paper, ‘Artificial Intelligence and Work: A review of the European Policy Landscape’, by Sally Williams, considers the risk-based approach of the EU AI Act and its capacity to protect workers from potential AI harms. It assesses key European policies concerned with the regulation of Algorithmic Management and AI at work paying particular attention to whether regulations address AI-driven monitoring and control systems, and the risks these systems can create for workers. It also identifies a number of areas where improvements are needed in the EU policy framework governing work-related use of AI. We thank our contributors for their interesting contributions and those who provided double blind reviews on their papers.
We are pleased to announce that in July 2025, the JIR Editorial Team formally adopted a Statement of Editorial Principles on Indigenous scholars and scholarship, grounded in the AIATSIS Code of Ethics, which articulates our responsibility to elevate Indigenous voices, challenge deficit narratives, uphold Indigenous data sovereignty, and ensure ethical, respectful, and reciprocal engagement with Indigenous communities. This statement is now available on the JIR website and is linked to Author Guidelines (see: News Update @ https://journals.sagepub.com/home/jir).
Finally, we also take this opportunity to welcome our two new JIR Early Career Researcher (ECR) Editorial Mentees to the team, Dr Eva Herman, Research Fellow, Decent Work and The City-Alliance Project, Manchester Business School, UK and Dr Vicente Silva, Researcher at Instituto de Investigación en Ciencias Sociales UDP, Chile and Research Fellow, Institute for the Future of Work, UK.
