Today the industry body, Digi, released the Australian Code of Practice on Disinformation and Misinformation.
AMAN also welcomes the improvements made in the Australian Code of Practice on Disinformation and Misinformation, picking up a number of recommendations that our organisation made in the draft stage. Specifically
- The definition of harm has been expanded to include democratic political and policymaking processes and public goods such as the protection of citizens’ health, protection of marginalised or vulnerable groups, public safety and security or the environment. However, the word ‘imminent’ needs to be deleted as ‘serious harm’ is adequate.
- The Code includes misinformation, where the harm is not clearly intended.
- The definitions of disinformation include behaviour driven by bots/AI as well as human and bot behaviour to manipulate conversations, as evident in online echo chambers.
- The amount of excluded content has been much better defined and limited.
- Protected news content has been better defined to mean news content that is the subject of a published editorial code which sets out content standards and or/complaints mechanisms.
- The Code now includes clearer objectives for adjudication, such as ‘Provide safeguards against Harms that may arise from Disinformation and Misinformation’ and ‘Disrupt advertising and monetisation incentives for Disinformation’.
The Australian Muslim Advocacy Network (AMAN) remains very concerned that
- The definition of harm still requires the harm to be ‘imminent’ which fundamentally misconstrues the reality of how disinformation creates serious harm over time. This may exclude vulnerable and marginalised communities affected by dehumanisation in online echo chambers from having recourse. How will imminent harm be demonstrated? If the word ‘imminent’ was removed, a serious harm would still need to be proven, which is reasonable and proportionate.
- The Code is still opt-in, and while it is required for signatories to comply with the first objective about preventing harm, and to provide a business case for why other objectives do not apply, it is not clear if a decision to grant exemption from those objectives will be judicable. It would have been better policy to have an opt-out process for smaller companies.
- The complaint process is not yet developed, and if it is being designed by the industry body, we are not sure how effective it will be.
- It is still unclear how independent the governance arrangements will be, and to what degree expertise in civil society, will be harnessed.
- The template for gathering data from signatories should be designed using expertise from independent research and civil society.