The Trump administration has fired a warning at Australia over the Albanese government’s sweeping new online safety plans, escalating an international dispute over how far governments should be able to regulate powerful US technology companies.
The warning followed the release of Australia’s draft Digital Duty of Care laws, which would require online services to take greater responsibility for preventing foreseeable harm and could expose companies to penalties of up to $109.2 million for serious non-compliance.
For social media users aged over 16, the government also wants platforms to provide a genuine choice over how their feeds operate. Under the proposed “My Feed, My Way” model, Australians could choose whether they want personalised recommendations selected by an algorithm or a default feed built around the friends, creators and accounts they have chosen to follow.
The proposal has now caught the attention of Washington.
White House spokesperson Kush Desai said President Donald Trump had warned US trading partners against imposing “digital services taxes, fines, and other forms of extortion” on America’s technology sector, adding that the administration would continue raising the issue with trading partners.
The remarks amounted to a clear warning to Canberra, although there is an important distinction: Trump has not been reported as personally making a specific public statement about Australia’s Digital Duty of Care proposal. The intervention came through his administration and reflects a broader White House campaign against foreign measures it believes unfairly burden major American technology businesses.
That distinction matters as debate intensifies in Australia over whether the new rules are principally an online safety measure or another front in an increasingly global confrontation between governments and US-based technology giants.
What Labor is proposing
The Albanese government released the exposure draft of the Online Safety Amendment (Digital Duty of Care) Bill 2026 on 8 September.
Rather than relying primarily on complaints after harmful material has already reached users, the proposed model would shift more responsibility onto online service providers themselves.
Platforms would be expected to identify foreseeable risks associated with their products, systems and features, take reasonably practicable steps to reduce those risks and document what they have done.
The approach extends beyond traditional social media. Communications Minister Anika Wells has said the duty is intended to reach online services including gaming platforms, apps, messaging services and AI chatbots where relevant.
For children, the government wants stronger protections against a series of psychosocial harms, including content promoting eating disorders, misogyny, pornography, the glorification of crime and material capable of causing serious distress.
Adults would continue to be protected in relation to illegal online material rather than being subjected to a new general government test of acceptable political opinion.
Wells has specifically said political misinformation is not included in the Digital Duty of Care proposal and has rejected claims that the legislation is designed to give the government power to decide which political opinions Australians can see.
Instead, Labor argues the core of the legislation is “safety by design”: making companies assess how the mechanics of their products may cause harm before those harms occur on a systemic scale.
Australians would be given a choice over algorithms
The most publicly visible part of the proposal is likely to be the algorithm choice.
Social media companies would be required to notify new and existing users and give them a choice about their default feed.
A user who wants personalised recommendations could continue using them. Someone who does not want the platform constantly selecting recommended posts, videos or accounts could instead choose a feed focused on people and creators they actively follow.
The government says the choice must be genuine and enduring rather than being buried deep inside settings or repeatedly overridden by a platform.
Albanese has characterised the proposal as a transfer of control from technology companies to individual users.
“This is not about giving government control, it is about giving people control,” the Prime Minister said when announcing the draft.
The proposal reflects growing concern about recommender systems that are designed to predict what will keep a person engaged and then serve increasingly personalised material based on their behaviour.
Supporters of reform argue those systems can amplify extreme, disturbing or addictive material because content that generates anger, fear or intense emotional engagement can also keep users scrolling.
Critics, however, warn that regulating how feeds operate could affect businesses, publishers, creators and political organisations that depend on algorithmic recommendations to reach people who do not already follow them.
Meta’s Instagram chief Adam Mosseri has also warned that a purely chronological or following-based experience can have its own shortcomings, including allowing brands and high-frequency posters to dominate feeds.
Under Labor’s proposal, Australians would not be forced to abandon algorithms. They would be given the option to choose them or reject them.
Maximum penalty is $109.2 million
The potential financial consequences for technology companies are substantial.
The government’s official announcement says failure to comply with the Digital Duty of Care could attract penalties of up to $109.2 million, with the independent eSafety Commissioner responsible for compliance and enforcement.
That figure is slightly higher than the rounded $100 million figure widely used in political and media coverage.
The government argues penalties of that scale are necessary because some of the companies being regulated are among the largest and wealthiest corporations in the world. Smaller penalties, ministers say, risk becoming little more than a cost of doing business.
The framework is intended to use graduated enforcement, with the most serious penalties directed at egregious or systemic failures rather than every individual incident of harmful content appearing online.
The concept was developed after the statutory review of Australia’s Online Safety Act recommended moving towards a preventative duty of care rather than leaving regulators and victims to respond after damage had already occurred.
Why Washington is pushing back
The US response cannot be viewed in isolation from the Trump administration’s wider policy towards foreign regulation of American technology companies.
Washington has repeatedly objected to digital taxes, regulatory penalties and other measures it believes discriminate against US firms or extract revenue from them for the benefit of foreign governments and industries.
Companies such as Meta, Google and other major digital businesses operate internationally but are headquartered in the United States, meaning disputes over their regulation can quickly move beyond domestic policy and into trade and diplomatic relations.
Trump has previously threatened retaliatory action when foreign governments impose measures his administration considers discriminatory towards American technology businesses.
That has created a difficult balancing act for Canberra.
Australia considers the United States its most important strategic ally, but successive Australian governments have also shown a willingness to regulate global technology companies in areas including online safety, payments to news publishers and children’s access to social media.
The latest proposal potentially brings those two priorities into direct contact.
Albanese government plays down threat to US relationship
The Australian government has nevertheless rejected suggestions the dispute will seriously damage relations with Washington.
Albanese has emphasised that Australia is a sovereign country entitled to make laws in its national interest, while also describing his relationship with Trump as very good.
Wells has similarly played down claims the legislation represents an attack on American companies.
Asked whether Australia risked consequences from Washington, the Communications Minister said she had not been approached by the US Consul General about the Digital Duty of Care during her time in the portfolio.
She said the government had the right to act to protect Australian parents and children.
The issue could nevertheless become more prominent as Albanese and Trump prepare for further high-level engagement, particularly because technology regulation has emerged as a growing source of friction between the United States and several of its allies.
European governments have faced similar pressure from Washington over regulation of major digital platforms.
The laws are not yet final
Despite the increasingly heated political rhetoric, the Digital Duty of Care is not yet Australian law.
The government has released an exposure draft for consultation and is accepting feedback until midday on 22 September.
That process gives technology companies, civil society groups, safety advocates, academics and other interested parties an opportunity to challenge or propose changes to the framework before final legislation is introduced.
The government says it intends to bring the legislation to parliament during 2026.
That means some details could still change, including how particular obligations are applied and how regulators deal with different categories of online services.
The proposal will also face a difficult political debate.
The Coalition has raised concerns about the breadth of the legislation and the powers potentially available to ministers and regulators. Opposition figures have warned that poorly drafted rules could create risks for freedom of expression even if the government says political speech is not the target.
Labor rejects that argument and says opponents should identify where political censorship actually appears in the draft.
The Greens, meanwhile, have pushed for tougher restrictions in some areas, including stronger default protections against algorithmic recommendations.
The disagreement means the final model could still be altered as the legislation moves through consultation and parliament.
Australia’s online safety experiment enters a new phase
The Digital Duty of Care follows Australia’s social media minimum-age regime, which prevents children under 16 from holding accounts on specified age-restricted social media platforms.
The government says more than five million accounts have been removed, restricted or deactivated since the age rules took effect, although debate continues about how effectively determined teenagers can bypass restrictions.
Labor now wants to move beyond age verification and address the way online products themselves are designed.
That represents a significant expansion of Australia’s approach.
The central question is no longer simply whether someone should be allowed onto a platform. It is increasingly about what responsibility a platform has for the systems that decide what users see after they arrive.
For Albanese, the argument is that companies exercising enormous influence over Australians’ daily lives should face safety obligations comparable in principle to other industries.
For the Trump administration, the growing international willingness to impose major penalties and regulatory burdens on US technology companies raises a different concern: whether foreign governments are using regulation in ways that unfairly target American economic interests.
Those competing positions have turned an Australian online safety proposal into something considerably larger.
If the draft becomes law in anything close to its current form, Australia could once again become an international test case for aggressive regulation of social media and digital platforms.
And with Washington now publicly signalling its concern, the debate over Australians’ feeds may increasingly become part of a much broader argument about sovereignty, trade and who ultimately sets the rules for the global technology industry.
