AI in the Australian Workplace: Your Rights and Employer Obligations in 2026

The Albanese Government has moved to put a national structure around artificial intelligence: a new Office of AI sitting within the Department of the Prime Minister and Cabinet, backed by a single national framework and Australian AI standards, alongside a National AI Plan and the Australian AI Safety Institute.

Crucially for employers and workers, the Government has decided not to introduce a standalone "AI Act" or blanket mandatory guardrails at this stage. Instead, Australia is regulating AI through the laws that already exist — employment, anti-discrimination, work health and safety (WHS), privacy and consumer law — supported by voluntary guidance and the new safety institute.

That's the headline most coverage misses: there is no single new AI law to comply with. The obligations that bite at work in 2026 come from existing rights being applied to AI, plus a handful of targeted reforms already legislated.

Two specific changes matter most for Australian workplaces this year.

1. Automated-decision transparency (Privacy Act). From 10 December 2026, organisations covered by the Privacy Act must spell out in their privacy policy where they use computer programs to make (or substantially help make) decisions that could significantly affect a person — including the kinds of personal information used. For employers, that reaches AI used in recruitment, rostering, performance and monitoring.

2. NSW Digital Work Systems Act. New South Wales passed the Work Health and Safety Amendment (Digital Work Systems) Act 2026, which places WHS duties on businesses that use AI, algorithms, automation or platforms to allocate and manage work. Employers must consider whether those systems create or amplify psychosocial hazards — think constant surveillance, loss of autonomy, or relentless algorithm-set targets.

Even without a dedicated AI law, you are not without protections when AI touches your job:

• An explanation. Where a substantially automated decision significantly affects you — a knock-back in hiring, a performance rating, a termination — you can reasonably expect a plain-English account of how it was reached, not "the system decided".
• Human oversight. Significant employment decisions (hiring, firing, promotion, discipline) should keep a real human in the loop. A fully automated sacking is legally fraught.
• Freedom from discrimination. An AI screening tool that filters out candidates on the basis of a protected attribute is still unlawful discrimination — the algorithm is no defence.
• Protection from unreasonable surveillance. Monitoring and algorithmic management that create psychosocial risk now sit squarely within WHS duties, especially in NSW.

If an AI-driven decision has cost you your job, our unfair dismissal checker can help you work out where you stand.

If your business uses AI or automated systems to hire, roster, monitor or manage people, four practical steps put you on the right side of the 2026 rules:

• Keep humans in the loop for significant decisions — document the human review step for hiring, dismissal and performance outcomes.
• Assess psychosocial risk from any system that sets targets, tracks activity or allocates work — this is now an explicit WHS duty in NSW and good practice everywhere.
• Update your privacy policy before 10 December 2026 to disclose automated decision-making, ahead of the Privacy Act deadline.
• Audit your AI tools for bias — recruitment and HR algorithms must comply with anti-discrimination law.

For a broader compliance view, see our employer hub and the 2026-27 compliance calendar.

Australia has picked a light-touch, existing-laws path on AI rather than a sweeping new statute. That's easy to read as "nothing changes" — but at work, the opposite is true. Transparency obligations, WHS duties for algorithmic management, and long-standing rights against discrimination and unfair dismissal all now apply to AI, and the first hard deadline (Privacy Act automated-decision disclosure) arrives in December 2026.

Whether you're a worker facing an algorithm or an employer deploying one, the safest move is to treat AI decisions the way the law already treats human ones: explainable, overseen, fair, and safe. Not sure how the rules apply to your situation? Ask our AI advisor — grounded in current Australian workplace law and data.