Australia’s national online safety regulator today published a coordinated set of measures to reduce harm from AI driven image abuse on social media, combining strengthened regulatory guidance, a targeted grants round for victim support and new compliance expectations for platforms operating in Australia. The package, announced by the eSafety Commissioner on October 5, 2026, responds to rising use of artificial intelligence tools that create or manipulate sexualised images and video, including so called nudify or deepfake services, and to ongoing concerns about under age access to mainstream social platforms. New regulatory guidance and compliance tests The eSafety office has issued updated guidance that clarifies how providers must treat AI generated sexual content and deepfakes that depict children or young people, and what reasonable steps platforms should take to prevent distribution and to quickly remove material when it is reported. The guidance sets out sharper evidence expectations for removal notices and explains how metadata, provenance and content moderation signals will be weighed in enforcement decisions. As part of these changes, the regulator is requiring platforms to adopt automated detection pipelines for suspected AI generated nudity that are tuned to minimise false positives, to keep auditable records of removal decisions, and to provide clearer, faster reporting routes for victims and for third party reporters. Platforms will be expected to supply the eSafety Commissioner with timely information when the office opens formal investigations. Grant funding to support victims and local services The announcement also includes a new targeted grants round to expand support services for people harmed by AI enabled image abuse. Grants will prioritise organisations that can provide rapid digital takedown assistance, trauma informed counselling, and legal support to survivors, with a particular emphasis on services that help children and teenagers and their families. The funding will also support capacity building for schools and community organisations so they can recognise AI based image abuse and respond safely when it appears on feeds and messaging services used by young people. Why the regulator acted now The eSafety Commissioner says the move follows continued reports of AI generated sexualised images circulating on mainstream social platforms, and that existing rules left gaps in how such material is detected, classified and removed. Regulators globally have struggled to keep pace with rapid advances in generative AI and with the proliferation of standalone services that produce manipulated images which are then shared across social networks and messaging apps. In Australia’s case, the intervention dovetails with the country’s world leading social media minimum age rules and recent legislative upgrades that increased the regulator’s investigatory powers. The office told platforms and developers it will take a more active enforcement stance where evidence shows systems or processes are inadequate and victims are not receiving effective redress. Industry response and platform obligations Major platforms and AI tool providers are now on notice that the eSafety Commissioner expects demonstrable technical and administrative safeguards to be in place. The updated guidance emphasises both front end measures, such as preventing features that make easy bare image generation of minors, and back end controls, such as transparent audit logs for moderation decisions. Regulatory officials indicated they will prioritise investigations where Australian children are directly harmed, where services used to create manipulated images are hosted or managed overseas but materially serve Australians, and where platforms fail to act despite repeat reports. What this means for Australians and creators For families and young people, the changes are intended to produce faster takedowns and clearer routes to get help. Schools and non profits receiving grant funding should be better equipped to advise young people and to work with parents when incidents occur. For content platforms and AI developers, the guidance signals a higher bar for compliance in Australia. Providers that cannot show they are taking reasonable steps to detect and remove AI generated child sexual content, or that do not cooperate with lawful information requests from the regulator, risk enforcement action under the Online Safety Act and related rules. Looking ahead The eSafety Commissioner said the office will monitor the effectiveness of the new guidance and grants, and will update standards as technology changes. Officials also said they will continue to engage with international counterparts to share technical approaches for identifying manipulated images, and to press for takedown cooperation from services that operate across borders. The package announced today is the regulator’s latest attempt to square rapid AI innovation with protections for children and vulnerable people online. Observers say real change will depend on strong technical implementation by platforms, timely reporting and resourcing for victim support services. The new funding and clearer enforcement expectations are intended to accelerate all three elements.