The Australian Data and Insights Association has commissioned an independent review of its Privacy (Market and Social Research) Code. It is the only code registered under the Australian Privacy Principles, and it has not kept pace with how data is actually being used.
Former Tasmanian senator Terry Aulich is leading the review. The focus areas are AI adoption, children and youth privacy, overseas data sharing and enforcement obligations. The review is now in its member consultation phase, timed to land ahead of ADIA's annual Leaders Forum on 7 May.
The timing is not accidental. From 10 December 2026, APP entities will be required to disclose in their privacy policies the types of personal information used in substantially automated decisions. That is less than seven months away. Any business using AI to segment audiences, score leads or personalise content will need to explain how those systems work in plain language.
Automated decision-making disclosure requirements take effect in Australia
The ADIA code was originally designed for traditional market research. Surveys. Focus groups. Panel data. The world it was written for has been replaced by one where customer data flows through AI models, third-party platforms and cross-border pipelines that did not exist when the code was drafted.
Why it matters
Most Australian marketers do not think about the ADIA privacy code. They should. It is the regulatory framework that governs how research data can be collected, stored and shared. When AI tools ingest customer data for analysis, they are operating within the scope of this code, whether the marketer using them realises it or not.
The review signals a broader shift. Privacy regulation in Australia is moving from consent-based frameworks toward transparency and accountability models. The question is no longer "did the customer click agree" but "can you explain what your system did with their data."
What to do about it
Audit your data flows. Map every point where customer data enters an AI system, whether that is a recommendation engine, a scoring model or a content personalisation tool. Document what data goes in, what decisions come out, and how those decisions affect the customer. When the new disclosure requirements land in December, you will need that map. Start building it now.
