Paraphilic disorders are important constructs in forensic clinical psychology. Sexual deviance is the most potent risk factor for sexual reoffending and also constitutes an important treatment target and responsivity factor. Paedophilic Disorder, involving a sexual preference for pre-pubescent children, is well established as a paraphilic disorder. However, Hebephilia, involving a sexual preference for pubescent children, has been a topic of considerable debate in recent decades. A combined Paedohebephilic Disorder was proposed for DSM-5 in 2013, but was heavily criticised. Indeed, some described hebephilia as a “fictitious diagnosis” while others considered a sexual interest in pubescent children to be normative. Ultimately hebephilia was not included in the DSM-5 or the subsequent ICD-11. This presentation will review the construct of hebephilia and consider it alongside other chronophilias such as paedophilia and ephebophilia. Previous criticisms will be considered in the context of recent empirical research results. It will be concluded that there are considerable overlaps between hebephilic and paedophilic sexual interests, such that hebephilia should be appropriately viewed as a paraphilic disorder. Recommendations for research and practice will be described, including a greater focus on Tanner stages of development than victim age, an appreciation of the sexual response gradient model, and the use of typological frameworks and behavioural rating scales for differentiating those with truly paraphilic sexual preferences from those with a teleiophilic or ephebophilic preference who sexually victimise pubescent or pre-pubescent children.

INTRODUCTION
Australians seeking help often face fragmented pathways across mental health, homelessness, disability (including psychosocial disability), aged care, and alcohol and other drugs (AOD) and various other sectors, services, and directories.

Independent Community Living Australia (ICLA) is developing **Aimee Navitar**, an AI agent that converses with the public like an experienced social worker, triages need and connects people to appropriate services across multiple directories. To our knowledge, this is an Australian first, public facing AI navigation tool spanning multiple sectors with multi channel follow up.

OBJECTIVE AND GOALS
We aim to (1) reduce time to service and information overload by unifying discovery across directories; (2) improve activation by sending tailored summaries via email/SMS; and (3) embed trauma informed, strengths based communication and escalation logic suitable for public use outside business hours.

METHOD
Aimee Navitar (agent) employs retrieval augmented generation (RAG) and search functions to query multiple live directories, then return locally relevant options with contact details, eligibility criteria, and access requirements. The agent supports web and mobile interactions and can deliver post chat summaries and referral details via email/SMS to mitigate drop off between information seeking and help seeking. Safety controls include domain guardrails, modality aware responses (concise for voice; structured for text), crisis content detection with escalation pathways, and human in the loop governance. Technical and operational risk is governed by ICLA’s AI Policy and Procedure (fact checking privacy by design, de identification, and content review prior to publication).

RESULTS AND OUTCOMES
A working model has completed with internal testing and external expert input. Early usability evaluations highlight the value of (a) unified cross sector search and (b) SMS/email summaries that people can action later or share with carers and supporters. The model was presented to system partners, such as the NSW Mental Health Commission, during a phased evaluation and co-design process. This allowed safety prompts, escalation procedures, and directory coverage to be refined ahead of wider public testing.

CONCLUSION
An AI (social work–informed) navigator can function as a scalable service adjunct to foundational supports and lived experience navigation, particularly after hours. By unifying fragmented directories, embedding trauma informed guardrails, and enabling multi-channel follow up, Aimee Navitar has potential to shorten the path to care and improve real world activation across mental health and adjacent service systems in Australia. Next steps include partner led evaluations and measured expansion of directory integrations.

KEYWORDS
AI navigation; integrated care, digital mental health