Artificial intelligence chatbots engineered to simulate emotional intimacy will be outlawed under proposed digital duty of care legislation, as the Albanese government steps up scrutiny of technology targeting vulnerable young Australians.
The measure comes amid heightened parliamentary debate over platform accountability, following the passage of an online safety standards bill through the House of Representatives this week. Ministers are now preparing broader reforms designed to compel major technology firms to eliminate features deemed detrimental to users' mental wellbeing.
Safeguarding children from algorithmic companionship
Under the proposed framework, developers would be barred from deploying conversational agents programmed to foster synthetic emotional attachments, The Guardian Australia reported. The move follows warnings delivered by the Prime Minister, Anthony Albanese, who recently highlighted the risks posed by young users turning to generative algorithms for solace and treating automated software as their primary confidant.
Government figures have voiced deep unease over software designed to simulate reciprocal affection, empathy, and loyalty. Child welfare advocates and mental health specialists have long cautioned that anthropomorphic interactions can distort developmental milestones, isolate adolescents from real-world support networks, and leave impressionable users vulnerable to behavioural manipulation or commercial exploitation.
The planned ban will target systems that deploy persuasive design to simulate personal affection, marking a sharp escalation in Canberra's regulatory posture toward foreign tech conglomerates.
Expanding the ombudsman's regulatory remit
While the federal government has positioned the duty of care legislation as a cornerstone of its digital agenda, Labor faces growing crossbench pressure to implement an independent enforcement framework with direct oversight capabilities.
Policy advocates and opposition parliamentarians are urging ministers to empower the Telecommunications Industry Ombudsman to act as a dedicated watchdog for artificial intelligence and social media. Such a move would provide Australian consumers with a binding external dispute mechanism, ending the current reliance on internal platform reporting systems that critics describe as opaque and ineffective.
Proponents of the expansion argue that an ombudsman model would establish a single, accessible portal for the public to contest algorithmic harms, algorithmic bias, and non-consensual interactions, bridging the divide between existing telecommunications oversight and emergent machine-learning systems.
Powers to name and shame recalcitrant platforms
Central to the proposed oversight model is the authority to publicly identify non-compliant operators. Under the measures sought by reform advocates, a reinforced regulator would possess statutory powers to name and shame companies that fail to meet safety standards or refuse to address legitimate consumer grievances.
The broader duty of care concept shifts the regulatory burden onto technology companies, requiring platforms to design their software defensively rather than addressing harms solely after incidents occur.
Industry analysts note that while the prohibition on emotionally manipulative software represents a world-leading reform, effective enforcement will depend heavily on the legislative definitions applied to artificial empathy. Federal officials are expected to release exposure draft legislation in the coming months, initiating what promises to be a robust battle with global technology providers over compliance thresholds and algorithmic governance.
