US criticises Australia's proposed algorithm opt-out laws as 'censorship'

The US has taken a swing at Australia’s plan to give social media users the right to opt out of algorithms, calling it “censorship of protected speech.” That is a sharp label for a draft law that is still on the table, but Washington is clearly treating it as more than a domestic tweak to app settings.

Under Australia’s proposed digital duty of care laws, tech firms could face hefty fines if they do not offer users a way to turn off algorithmic feeds. The US embassy in Canberra made the criticism in a rare submission, saying it had “serious concerns” and warning that vague definitions of “harm” could open the door to “viewpoint-based censorship.”

The American submission also pushed on a second pressure point: how platforms would handle sensitive or controversial material. It argued that the rules could push companies to “pre-emptively demote” independent journalists or other voices, and warned that mandated design changes for algorithms could shape what people “see, say, and hear” beyond Australia’s borders as well. There was also a broader complaint that the plan could hurt Australia’s reputation as a place that supports innovation.

Anthony Albanese has tried to frame the proposal very differently. Speaking in New York while on a visit to the US for the UN General Assembly, he said the laws were not about giving government control but about giving individuals control over what they receive on their devices. The clash lands in the middle of a wider Australian push to police online harms, less than a year after the country introduced a social media ban for children under 16.

And then there was Donald Trump at the UN, doing his own thing and brushing aside AI guardrails entirely, while suggesting the word “artificial” makes it sound fake. Different script, same basic argument: politicians love control until somebody suggests controlling the platforms.