Item Catalogue
AI governance, regulation, strategy, and practice developments from monitored sources.
- Week of 15 December 2025
-
Tue 16 Dec 2025 Celebrating innovation at GovHack 2025 Government Finance (Department)
- Department of Finance sponsored GovHack 2025, setting challenges around service access and regulatory simplification.
- Winning teams built AI-powered platforms to help navigate government services and simplify small business compliance.
- Winners are invited to showcase solutions at the AI CoLab in 2026, linking hackathon outputs to APS capability development.
-
Australia earns global A-rank and Top 5 spot in digital transformation
- Australia ranked 5th globally in the World Bank's 2025 GovTech Maturity Index with a score of 98.5%.
- AI governance is mentioned as a contributing enabler, citing the APS AI Plan and responsible use policy.
- The item is primarily a digital transformation achievement announcement; AI is one thread among many.
-
New AI-powered tool to identify threats in space and improve national security
- UK researchers won a £610,000 grant to develop an AI tool for identifying threats in space.
- The Alan Turing Institute and University of Birmingham are leading the national security-focused project.
- Limited direct relevance to Australian federal agencies; included for context on AI-enabled space security research.
-
Apply on USAJobs: Open CAISI Position for an AI Research Scientist
- NIST CAISI is recruiting an AI Research Scientist focused on AI evaluation methods and trustworthiness assessment.
- The role signals continued US investment in AI measurement science and pre/post-deployment evaluation infrastructure.
- Job postings are low-signal for APS readers; this item is context only, not a priority for action.
-
Secure Software Development Framework (SSDF) Version 1.2 is Available for Public Comment
- NIST has released SSDF Version 1.2 for public comment, closing 30 January 2026.
- The framework addresses secure software development practices across the full software lifecycle.
- This is a general software security standard - AI is not the subject, limiting APS AI relevance.
-
Securing Smart Speakers for Home Health Care: NIST Offers New Guidelines
- NIST has finalised cybersecurity and privacy guidelines for smart speakers used in home telehealth settings.
- Guidelines draw on NIST CSF 2.0, Privacy Framework, and IoT baseline standards - no direct APS mandate.
- Limited direct relevance to Australian federal agencies; this is a niche US health-IoT security publication.
-
Just Published! Final NIST Telehealth Smart Home Integration Cybersecurity White Paper
- NIST finalises a cybersecurity white paper on smart home and telehealth integration risks, including IoT devices.
- AI is not the subject; focus is on cybersecurity and privacy risks in hospital-at-home deployments.
- Limited direct relevance to Australian federal AI governance - this is a US healthcare cybersecurity document.
- Week of 8 December 2025
-
AI Policy and Governance Newsletter — December 2025
- Australia's AI Safety Institute announced with $30m funding, beginning operations in early 2026.
- National AI Plan and APS AI Plan 2025 released, taking a guidance-over-guardrails approach - both directly relevant to APS practitioners.
- ANU survey found 77% of Australians rate AI attacks on people and businesses as a major or moderate threat - the top-ranked national security concern.
-
New Cloud Policy: Accelerating secure, modern government services
- DTA has released a whole-of-government Cloud Policy taking effect 1 July 2026 for non-corporate Commonwealth entities.
- Policy explicitly positions cloud as an enabler of AI and emerging technologies across the APS.
- Five core requirements cover cloud prioritisation, responsible adoption, security, cost transparency, and workforce capability.
-
CAISI Evaluation of Kimi K2 Thinking
- CAISI evaluated Kimi K2 Thinking, finding it the most capable PRC AI model at release but behind leading US models.
- The model is heavily censored in Chinese but relatively uncensored in English, Spanish, and Arabic - a notable deployment risk signal.
- CAISI's systematic benchmarking of PRC frontier models sets a precedent relevant to Australian agencies assessing open-weight AI procurement risk.
-
E-Leaders explore human centred design, measuring what matters, and rethinking investment
- DTA hosted OECD E-Leaders to discuss digital government delivery including AI measurement and digital identity.
- Only a quarter of surveyed countries conduct thorough AI impact assessments; Australia's Investment Oversight Framework was cited as a comparative example.
- AI governance is one thread among broader digital government topics; limited new actionable detail for APS AI practitioners.
-
Import AI 437: Co-improving AI; RL dreams; AI labels might be annoying
- Facebook researchers propose 'co-improving AI' - human-machine collaboration toward superintelligence - as safer than self-improvement.
- A commentary on EU AI labelling complexity warns that even simple policy ideas carry significant compliance costs.
- SimWorld, a high-fidelity RL training simulator, is released by multi-university researchers - limited direct APS relevance.
- Week of 1 December 2025
-
AI Policy overhauled with new Impact assessment tool and Procurement guidance
- DTA releases updated AI policy, a new AI Impact Assessment Tool, and AI procurement guidance, effective 15 December 2025.
- Updated policy mandates impact assessments for all AI use cases and requires agencies to develop and communicate a strategic position on AI adoption.
- An AI Review Committee for high-risk use cases across the APS is being finalised, with terms of reference expected in Q1 2026.
-
New guidance helps Australians identify AI-generated content
- The National AI Centre has released guidance on labelling, watermarking, and metadata for AI-generated content.
- Guidance targets businesses but applies equally to APS agencies producing AI-assisted public-facing content.
- Framed around reducing regulatory and reputational risk, not a mandatory standard - voluntary best practice only.
-
Repository Update: December 2025
- MIT AI Risk Repository Version 4 now contains over 1,700 coded AI risk categories drawn from 74 frameworks.
- Newly added frameworks include UK DSIT frontier AI risk analysis and MITRE chatbot risk taxonomy with mitigations.
- A structured, openly accessible reference - useful for agencies benchmarking their own AI risk classification work.
-
Research explores risks of using AI in the financial sector
- Alan Turing Institute researchers have identified new AI-specific risks emerging in financial sector applications.
- Financial sector AI risk findings are broadly transferable to Australian regulators and agencies overseeing fintech and financial services.
- Extracted text is truncated - full scope and specific risk findings cannot be verified from available content.
-
New AI model could enable real-time maritime surveillance onboard satellites
- Alan Turing Institute researchers have developed an AI model enabling real-time maritime surveillance processing onboard satellites.
- On-orbit processing could reduce latency in detecting vessels, with potential border security and defence applications.
- Extracted text is truncated; specific technical details, performance benchmarks, and limitations are not available for assessment.
- Week of 24 November 2025
-
Iason Gabriel: Value Alignment and the Ethics of Advanced AI Systems
- Google DeepMind philosopher Iason Gabriel discusses value alignment, democratic civility, and ethics of advanced AI assistants.
- Topics include aligning LLMs with human values, AGI and social power, and distributive justice - relevant to APS AI ethics frameworks.
- A research interview rather than policy guidance; useful for background understanding, not direct APS application.
-
Import AI 436: Another 2GW datacenter; why regulation is scary; how to fight a superintelligence
- OSGym enables cheap, scalable training of AI agents to operate computers across multiple applications.
- RAND paper models options for neutralising a rogue superintelligence - all carry severe collateral risks.
- Regulation commentary is US-focused and editorial in nature; limited direct APS policy signal.
- Week of 17 November 2025
-
AI Adoption: Built on trust, people, and tools
- The APS AI Plan launches with three pillars — Trust, People, and Tools — as a joint DTA, Finance, and APSC initiative.
- Agencies must appoint Chief AI Officers, designate accountable officers per use case, and maintain internal AI registers.
- A new AI Review Committee managed by DTA will provide cross-government scrutiny of high-risk AI use cases.
- Week of 10 November 2025
-
Fission for Algorithms: The Undermining of Nuclear Regulation in Service of AI
- AI energy demands are driving pressure to fast-track nuclear deployment, raising serious safety and regulatory concerns.
- LLMs are being proposed and deployed to generate nuclear licensing documents, with governance implications for high-stakes AI use.
- Australia has no direct nuclear power context, limiting immediate APS relevance beyond AI energy and governance themes.
- Week of 3 November 2025
-
AI Policy and Governance Newsletter — October/November 2025
- New polling shows 65% of Australians believe AI creates more problems than it solves, up 8 points since 2023.
- Australia fell from 15th to 23rd in the IMD World Digital Competitiveness Ranking, with AI policy rankings dropping from 8th to 34th.
- Treasury's ACL review found existing laws broadly adequate for AI harms, while eSafety registers mandatory codes for AI chatbots effective March 2026.
-
Democratising environmental forecasting in the age of AI
- Alan Turing Institute blog explores AI-powered environmental forecasting tools and their potential to protect lives.
- Extracted text is a teaser only - no substantive findings, data, or policy recommendations are available.
- Limited direct relevance to APS AI governance; low signal due to minimal content extracted.
- Week of 27 October 2025
-
The UK’s critical infrastructure is at risk from cyber-attacks. Our AI tools will provide a new line of defence
- The Alan Turing Institute is developing AI tools to defend UK critical national infrastructure from cyber-attacks.
- AI-augmented cyber defence for critical infrastructure is a growing concern for Australian agencies managing similar systems.
- Extracted text is minimal; substantive detail on methods or findings is not available from this item.
- Week of 20 October 2025
-
Supporting safer AI adoption: updated guidance for Australian business
- NAIC has released updated Guidance for AI Adoption, consolidating the Voluntary AI Safety Standard into 6 key practices.
- Guidance includes practical tools - AI policy template and AI register template - usable by agencies and businesses alike.
- Primary audience is Australian business, but the frameworks and templates are directly applicable to APS AI governance work.