Australia: Technology

The recent Hugging Face/OpenAI incident reveals critical lessons about AI agent security and confidential information exposure. When AI systems gain access to internal repositories containing source code, patent materials, and trade secrets, businesses face unprecedented risks that require deliberate controls and IP-focused governance before deployment.

The Office of the Australian Information Commissioner (OAIC) received 1,205 data breach notifications in 2025, the highest number since Australia's Notifiable Data Breaches (NDB) scheme commenced in 2018. The figures highlight an environment increasingly shaped by persistent cyber threats and AI-enabled risks.

Following the Aristocrat Technologies v Commissioner of Patents decision and updated IP Australia guidance, this article examines strategic approaches to claiming AI and machine learning inventions in Australia. It explores how to identify where real value lies in AI systems, support multiple claiming pathways across the AI lifecycle, and draft patent applications that withstand characterisation challenges by demonstrating practical implementation rather than abstract results.

The general rule is that a business cannot send unsolicited emails to market, sell, and promote their goods and services.

"Statement Expectations" sets out ACMA views on the minimum requirements needed to comply with the Spam Act.

This heavy penalty was a reminder to all businesses to ensure that their marketing complies with regulations.

A prisoner made an application to Queensland Corrective Services seeking to have some of her eggs extracted and frozen so that she may undergo ‘assisted reproductive technology’ upon her release from imprisonment.

This article examines the rise of AI wearables—particularly AI-enabled smart glasses—in the workplace, covering both employer-driven use cases (such as logistics, healthcare, and remote operations) and employee-driven use (including accessibility, personal devices brought to work and covert recording).

The Personal Injury Commission of New South Wales has issued its first decision addressing the use of generative AI in legal proceedings, establishing clear expectations for disclosure and verification when AI tools are employed. In Sentinel Community Services Pty Ltd v Clark Chesaites, an employer's undisclosed use of AI resulted in incorrect case citations and irrelevant statutory references, prompting the Commission to emphasize that compliance with AI disclosure rules applies equally to self-represented

A new enterprise agreement between Private Media & the Media, Entertainment & Arts Alliance, approved by the FWC restricts the use of AI in editorial work.

This draft legislation is ambitious in scope and represents Australia's answer to comprehensive crypto-asset reforms.

OAIC's recommendations on managing records of messaging apps provide useful lessons and practical solutions.