Daniel Mulino (Fraser, Australian Labor Party) Share this | Hansard source

I agree with some key aspects of the previous contribution—in particular, the importance of this issue, the importance of protecting children and the importance of steadily working through the eSafety Commissioner's recommendations. Where I disagree with the previous contribution is in its misrepresentation, in my view, of the minister's position when it comes to her response to that report. I will step through that carefully during my contribution.

The Roadmap for age verification was released in March 2023, and the government responded in August 2023—and I am going to provide a couple of brief quotes from that response later on in my contribution. I want to make the observation that there were some high-level guiding principles for this age-verification report, an important report and a very thoughtful report:

1. take a proportionate approach based on risk and harm

2. respect and promote human rights

3. propose a holistic response …

4. ensure any technical measures minimise data and preserve privacy

5. consider the broader domestic and international regulatory context

6. consider what is feasible now and anticipate future environments.

There were two key elements to a legislative or regulatory environment as put forward in that road map. One was establishing expectations and requirements for service providers to apply age assurance and other complementary measures, but also that there be established a regulatory scheme for the accreditation and oversight of age assurance providers to promote privacy, security, strong governance, transparency, trustworthiness and fairness. I want to add that, when you read this report, it's absolutely clear that the commissioner was recommending a holistic approach, reflecting that children access pornography and other inappropriate material in both intentional and non-intentional contexts.

I want to put to the chamber that this is a very complex issue when it comes to actually implementing a measure that, if it were able to be implemented in a simple way, would provide protection. But there are a raft of considerations that need to be taken into account, and many of these were explicitly identified in the road map. Age assurance measures—and this is something directly referred to in the road map—have the potential to deter users from accessing compliant sites, which may lead to situations where they instead follow what you might call a path of least resistance and end up on high-risk sites. There are also a whole range of privacy, security and reliability issues.

For the sake of putting this issue in context, I will clarify terminology. Age assurance is what is generally referred to as what we are seeking here, which is a combination of two different approaches: age verification, which could potentially be achieved through physical or digital identification; or age estimation, which would be achieved through a range of techniques, including estimating the age of the person by using facial images, voice or any number of other techniques. That second set of processes has a range of accuracy, but, as the previous speaker identified, the accuracy of age estimation using certain techniques is increasing. But it is also the case that using certain age estimation techniques does raise privacy issues, which would need to be managed through a range of very strong governance procedures.

It's important to note that the road map suggested a pilot for the reason that there are many issues that need to be managed as we move towards an age assurance system. Even the road map didn't say that we should immediately move towards a mandatory system. This is something that clearly needs to be piloted and worked through. The recommendations from the road map indicated that there should be funding of specialist researchers and that there should be a pilot—and that's exactly what the government is currently considering. The pilot is under consideration, and scoping work is currently being undertaken through a range of departments.

See this speech in context