Australia's House of Representatives passed a sweeping online safety bill on Wednesday that would force social media platforms, online games, apps and AI chatbots to keep their users safe by law — and fine companies up to A$109.2 million if they fail. The Online Safety Amendment (Digital Duty of Care) Bill 2026 places what the government calls "an overarching obligation" on tech platforms "to provide a safe online environment for all Australians," replacing the existing patchwork of industry-written codes and standards with a single legal duty of care.
The change at the heart of the bill is who carries the burden. Instead of parents and users having to chase harmful material after it appears, companies would have to identify the risks on their own services, carry out regular written risk assessments, and act to stop the harm before it happens. Prime Minister Anthony Albanese told Parliament the onus is not on children or parents to act — it is on the companies to act to make sure their services are safe.
Communications Minister Anika Wells introduced the bill on October 15, and it cleared the lower house in under two weeks. The night before the vote, Albanese made an emotional appeal to the chamber, saying "the stories we hear about people identifying a chatbot, a non person as their best friend, should be a huge wake up call for every member of this Parliament to be prepared to act." He again implored the opposition Coalition to consider voting for the bill.
Children sit at the centre of the new duty. Services would have to protect everyone under 18 from harmful material, conduct and design features — including pornography, content that glorifies crime or life-endangering stunts, content that encourages eating disorders, content that promotes hostility towards women, and material that causes serious mental distress, including abuse and bullying. Addictive features that keep children hooked would also be captured.
The sharpest child-protection measure was added on the day of the vote itself. The government accepted amendments from independent MP Kate Chaney that put AI chatbot behaviour on the list of harms, requiring platforms to ban "material or conduct that simulates emotional intimacy and encourages or fosters a child's dependence on it in place of human relationships." In plain terms, a chatbot designed to act like a child's close companion — the stand-in friend the Prime Minister warned about — is exactly what the law is aimed at.
The concern is not theoretical. A 2026 survey of 1,950 Australian children aged 10 to 17 found 78 percent had used an AI assistant and 8 percent had used an AI companion; among children who had used either, 54 percent said they used the tools for personal or social reasons, and 20 percent reported potentially inappropriate or harmful interactions.
Adults get something from the bill too. Under the government's "My Feed, My Way" plan, social media services must let users aged over 16 choose whether their main feed is the algorithm's recommended content or simply the people and pages they follow. Another accepted Chaney amendment requires that choice to be "accessible, easy to use" and "prominently available," after concerns the platforms might bury it in a settings menu where nobody would find it.
The duty has real teeth. Breaching it carries a civil penalty of 60,000 penalty units — up to A$109.2 million. Australia's online safety regulator, the eSafety Commissioner, is responsible for compliance and enforcement, and the bill strengthens the regulator's hand: removal notices for apps and websites built to generate fake nude images, a halving of the time platforms get to respond to cyberbullying complaints from 48 hours to 24, minimum standards for complaints processes, and powers to run undercover checks on whether companies are complying. The duty itself would start 12 months after the bill becomes law, when the 17 existing industry codes are folded into the one framework.
Not everyone in the chamber was persuaded. The Coalition opposition voted against the bill, alongside One Nation. Opponents accused Labor of building an "Orwellian" system of censorship; opposition communications spokeswoman Sarah Henderson called it "a blank cheque for political censorship." Wells strongly denied the charge, saying on radio Wednesday that if the opposition really cared, the test was simple: "bring an amendment, not a slogan." The tech industry's own lobby group, which represents companies including Meta, Snapchat and Google, defended algorithms as helpful to users, while welcoming the discussion.
The argument reaches well beyond Australia — and straight to the United States. The plan has already drawn a rebuke from the Trump administration, which said Australia's new regulatory baseline on U.S. technology companies risks spreading overseas. Albanese, for his part, told the House the bill means that "rather than someone in the United States deciding what they see, they get to decide themselves." Wells has pointed to America as the alternative path: in August, Meta agreed to pay up to $18 billion in landmark settlements resolving social media claims from U.S. states. She argues a set of globally agreed safety standards would be a better outcome than negotiated settlements country by country, saying it is going to take "a coalition of lawmakers working together across the globe" to take the platforms on.
What happens next is the Senate, where the government does not hold a majority and the bill will be debated in the coming weeks. The Greens support a digital duty of care and hold the votes Labor needs, but they want the law strengthened — higher penalties, possibly tied to a company's global revenue, and a right for Australians harmed online to pursue private legal action and compensation. A Senate inquiry into the laws is due to report in late November, and the government hopes to pass the bill by the end of 2026, though its path through the upper house remains unclear. One idea that failed in the lower house — independent MP Sophie Scamps's proposal to expand the telecommunications ombudsman into an external complaints watchdog for big tech — will be pushed by the Greens in the Senate negotiations; the ombudsman, Cynthia Gebert, says her office would be well placed for the role and to "name and shame" companies behaving badly.
For ordinary families, the promise on the table is straightforward: the apps, games and chatbots their children use every day would have to be safe by design, not safe only after someone complains — and every Australian would get a genuine say over what fills their feed. Whether that promise survives the Senate is now the fight to watch.
📰 ProfiloPolitics Member News — choose your news, share it, earn credit: https://white-curlew-963569.hostingersite.com
Join ProfiloPolitics on Telegram
Get the news as it breaks — follow our Telegram channel for every story, first.
Follow @profilopoliticsnews