The US government has asked Australia to abandon its plan to make internet platforms use algorithms to suppress content based on "ambiguous government-defined safety criteria."
"We ask that Australia withdraw any plans to mandate that platforms algorithmically suppress content based on ambiguous government-defined safety criteria," the US said in its submission to the public consultation on the Online Safety Amendment (Digital Duty of Care) Bill 2026.
The consultation closed on September 22. The proposed legislation aims to introduce a duty of care that would make platforms protect Australians from material designated as harmful.
The US, in its submission, said that the Australian Online Safety Act censorship law of 2021 already provides "robust protections against unprotected, illegal content" and gives the eSafety Commissioner "significant takedown powers." The new duty of care would be another layer on top of that.
However, the US has "serious concerns" about the new rules, and believes that they may result in viewpoint-based censorship. "A framework that empowers governments to enforce vague definitions of 'harm' risks becoming a mechanism for viewpoint-based censorship, as platforms will be incentivized to over-moderate speech that should be protected to avoid regulatory penalties," the submission reads.
The US also fears that the new rules would hit not only users in Australia, but also those abroad, including in the US. "Mandated platform design features, especially when applied to algorithms, may affect what users see, say, and hear not just in or from Australia, but globally," the submission said.
The US is not happy with the idea of Australia imposing a "duty of care" on platforms that would require them to assess the risk of "harm" and take steps to mitigate it. This would apply to algorithms that power recommendation systems, logged-in features, endless feeds, feedback mechanisms, and time-limited features. The Communications Minister would be able to designate more types of harmful material, conduct, or design features, while the eSafety Commissioner would be able to order platforms to change their systems and processes.
The US is also concerned that the proposed rules would result in "rigid, one-size-fits-all platform design requirements" such as forced chronological feeds or opt-in recommendation systems.
The US would like Australia to define "foreseeable harm" and explain how this would work in practice, and who would make these assessments, and how they would be appealed.
The US also believes that the proposed rules could negatively affect independent journalists and others whose content might be considered controversial or sensitive, as platforms may decide to preemptively suppress it in order to comply with the new rules.
This would put platforms in a difficult position, as they would have to make a guess, and if they guess wrong, they would face penalties. If, however, they err on the side of caution and suppress content, they would not be punished for that.
The US government also said that "if Australia compels platforms to aggressively filter content in algorithmic feeds to satisfy a government mandate, and such filtering draws in protected speech, the effect on a disfavored speaker or media source's reach would not be meaningfully different than outright removal."
And, the US warned, "the U.S. government would likely consider such a measure as facilitating censorship."
The US also noted that the new rules could have a global impact, and that "the U.S. government has made clear its opposition to foreign measures that result in extraterritorial censorship of protected speech by Americans online."
