In a decisive move to curb online harm, Australia has doubled the maximum penalty for social media platforms that breach its Online Safety Act, raising the fine from A$555,000 to A$1.1 million per violation. The amendment, passed through parliament this morning, targets platforms that fail to remove violent or abusive content within 24 hours of notification. The escalation comes as Australia’s eSafety Commissioner, Julie Inman Grant, reported a 40% increase in complaints over the past year, with a surge in child exploitation and terrorist material.
This unilateral action by Canberra sits within a broader international context. Britain, long a champion of platform accountability, is spearheading a coordinated global effort to enforce stricter standards. The UK’s Online Safety Bill, currently in its final stages in the House of Lords, proposes penalties of up to 10% of global turnover for non-compliance, a move that has already prompted tech giants to reconsider their moderation policies. In parallel, the UK government has convened a meeting of the Five Eyes intelligence alliance to discuss cross-border mechanisms for holding platforms accountable.
The scientific community, including myself, views this as a positive but insufficient step. The physics of information dissemination shows that unregulated platforms act as amplifying cavities for falsehoods and hate. The doubling of fines does not change the underlying dynamics if enforcement remains incomplete. We have seen in climate science that delayed action leads to irreversible tipping points. Similarly, social media can face a tipping point where algorithms prioritise engagement over truth. The UK’s push for algorithmic transparency and real-time auditing could be the equivalent of a global carbon pricing mechanism for the information sphere.
Australia’s move is a response to the 2023 decision by the Federal Court that found a platform liable for hate speech after a single user’s post led to a violent protest. The court ordered the platform to pay damages, a ruling that has since been upheld on appeal. The new penalty structure aims to ensure that such liability is not just a cost of doing business. However, critics argue that the maximum fine is still too low compared to the platforms’ quarterly revenues, which for some exceed $10 billion.
Britain’s leadership is more structural. The UK is pushing for a multi-lateral treaty on platform accountability, similar to the Paris Agreement, but focused on digital rights and safety. The proposed treaty would include mandatory impact assessments for algorithmic systems, independent audits, and cross-border enforcement. The UK’s Secretary of State for Culture, Media and Sport has stated that “freedom of expression cannot be a cover for violence,” and that platforms must demonstrate how they protect users, especially children.
From a thermodynamic perspective, the internet is a high-entropy system. Without external constraints, it tends toward disorder. Regulations act as information content, reducing entropy by forcing structure. The doubling of penalties increases the energy required for non-compliance, but effective governance requires multiple reinforcements: transparency mandates, algorithm audits, and global cooperation. The UK’s approach is akin to building a comprehensive energy policy that combines carbon pricing with efficiency standards and innovation funding.
The immediate effect in Australia is likely to be limited. Platforms may shift their compliance resources to higher-risk areas, but the ability to identify and remove content within 24 hours is technologically feasible. Cloud computing and AI-based monitoring can flag content in milliseconds, but platforms must invest in human review layers to avoid over-censorship. The UK’s algorithmic transparency demands could unlock the black box of content moderation, allowing researchers to analyse bias and effectiveness.
Biologically, we can think of platforms as ecosystems. Introducing a predator, such as strict penalties, can control the population of harmful species, but it may also disrupt beneficial interactions. The key is to maintain ecosystem health through multiple mechanisms: fines, transparency, and user education. Australia’s doubling of penalties adds a predator, but the UK’s push adds a monitoring system and a global treaty that could establish a stable equilibrium.
In the long term, the cost of non-compliance must exceed the profit from engagement. The current fine doubling is a step, but until global turnover-based penalties become standard, platforms will treat fines as a business expense. Britain’s leadership is promising because it addresses the systemic issue. As a climate and science correspondent, I see parallel with the early days of carbon regulation: initial voluntary measures fail, then national penalties, and finally global treaties. We are in the national penalty phase for social media. The UK is attempting to skip to the treaty phase.
For the individual user, the impact is gradual. Australian users may see faster removals of harmful content, but the underlying algorithms that promote sensationalism will remain until transparency mandates force changes. The biosphere of information will not recover overnight, but the trajectory is encouraging. Data density here is low because the regulations are recent, but initial enforcement data from the Australian eSafety Commissioner shows a 15% reduction in reported harmful content within three months of previous fine increases. This is a statistically significant trend, but not yet transformative.
To summarise, the doubling of penalties in Australia is a necessary but insufficient step. Britain’s global push is where real change lies. The physics of social media requires a phase transition from self-regulation to external oversight. These legal frameworks are attempting to induce that transition. They may not be perfect, but they are the best tools we have to counter the entropy of online harm.
