Humane AI is an independent initiative developing practical frameworks to ensure artificial intelligence systems are fair, accountable, transparent, and aligned with human rights — particularly where decisions affect individuals and communities.
Humane AI is an independent, not-for-profit initiative focused on the responsible development and deployment of artificial intelligence systems in Australia.
The platform provides practical standards, policy guidance, and use case frameworks to support ethical and accountable AI adoption across high-impact sectors where automated decisions affect people's lives.
Sectors where AI systems increasingly shape rights, services, and public trust.
Tenancy assessment, allocation models, and eligibility scoring.
Credit assessment, fraud detection, and automated lending decisions.
Welfare, health, and social service decision-support systems.
Automated decision-making, administrative review, and public-facing AI services.
Predictive systems, risk tools, and AI-assisted investigations.
The Humane AI Standard provides a structured framework for assessing and governing AI systems based on six core principles.
Ensuring AI systems do not produce discriminatory outcomes.
Making AI decisions understandable and visible.
Establishing clear responsibility for outcomes.
Ensuring humans remain in control of critical decisions.
Protecting privacy and ensuring responsible data use.
Applying proportional controls based on impact.
Humane AI applies its framework to real-world scenarios to identify risks and recommend safeguards.
Automated systems used to assess borrower risk and lending decisions.
View AssessmentAI systems used to forecast crime risk and allocate policing resources.
View AssessmentGovernment systems used to identify compliance issues, eligibility concerns, or potential fraud.
View AssessmentCommentary on responsible AI, governance, and emerging technology issues.
How direct and indirect bias enter automated decision systems, and the practical measures organisations can take to identify, test, and mitigate discriminatory outcomes.
Read MoreMeaningful human oversight is more than a reviewer clicking approve. This piece examines what effective oversight looks like in high-impact AI systems.
Read MorePublic institutions face a unique transparency obligation. A review of disclosure practices, plain-language explanations, and public transparency registers.
Read MoreRisk-based approaches align controls with impact. An overview of how Australian organisations can classify, assess, and govern AI systems proportionately.
Read MoreAI systems are increasingly being used to inform decisions that affect individuals' rights, access to services, and participation in society.
Without appropriate safeguards, these systems can introduce bias, reduce transparency, and create unintended harm.
Humane AI exists to ensure these technologies are deployed responsibly and in alignment with public trust.
Humane AI promotes transparency and accountability in AI deployment through:
We welcome collaboration from policymakers, technologists, legal professionals, and researchers.