Letting users switch off social media algorithms is right — but the law must be smart

The idea that Australians should be able to switch off social media algorithms is simple to state and, at first glance, commonsense. The government’s proposal that firms such as Meta, Google and TikTok should face “substantial” penalties if they do not give users the option turns that principle into enforceable policy. But while the headline is welcome, getting the details right will determine whether such a law protects citizens or produces unintended harms. Why this matters to ordinary Australians is not hard to see. Algorithms determine what billions of users see every day: news, opinions, entertainment and advertising. They curate feeds, amplify content that keeps people engaged and shape the civic conversation in subtle — and sometimes corrosive — ways. The option to switch them off would give users more control over their online experience, potentially reducing algorithm‑driven polarisation, attention-grabbing misinformation and addictive pathways particularly harmful to young people. Yet the practicalities are thorny. Not all algorithms are the same. Some are overt ranking systems that personalise newsfeeds; others power search results or recommend content across platforms. A blunt requirement to “switch off algorithms” could undermine features many small businesses, creators and charities rely on to reach audiences without large advertising budgets. It could also push platforms to make simplistic toggles that look like choice but bury other mechanisms that continue personalisation in less transparent ways. The minister’s insistence on “substantial” penalties indicates seriousness about compliance. That is necessary because tech giants have shown they will only change where non‑compliance carries real cost. But penalties alone are not enough. The law must be precise about what constitutes an opt‑out, what forms of personalisation are covered, how users will be informed and how platforms must demonstrate compliance. A credible enforcement regime requires technical understanding inside government and durable mechanisms for auditing platform behaviour. Consumer choice must be meaningful. That means default settings should favour the least intrusive experience, clear explanations about what switching off an algorithm does (and does not) change, and protections for vulnerable users. It also means regulators should guard against perverse outcomes, such as platforms consolidating user‑facing control while moving personalisation deeper into ad targeting systems. There is a broader civic dimension as well. Democracies depend on a public square where information is not warped by hidden commercial incentives. Giving Australians the power to opt out of algorithmic curation strengthens individual autonomy and public deliberation. But to realise that potential, lawmakers need to combine principled intent with technical nuance, resourcing and independent oversight. The proposal to require an opt‑out and back it with penalties is a welcome step towards rebalancing power between platforms and users. Now comes the harder work: designing rules that deliver genuine choice, preserve legitimate benefits of personalisation, protect small voices and ensure regulation keeps pace with technology. If that’s done, Australians will have a more honest and humane digital public sphere. If it is not, the reform could be little more than a cosmetic fix.