Item Catalogue
AI governance, regulation, strategy, and practice developments from monitored sources.
- Week of 9 March 2026
-
New research highlights risks from state-sponsored hostile AI collaboration
- Alan Turing Institute report calls for attention to national security risks from hostile state-sponsored AI collaboration.
- Adversarial AI collaboration risks are directly relevant to Australian government AI procurement and partnership decisions.
- Extracted text is truncated - full report detail unavailable; assessment is based on limited source material.
-
Fri 13 Mar 2026 Data and Digital Ministers Meeting advances national priorities Government Digital ID, Finance (Department)
- The February 2026 DDMM agreed that emerging technologies, including AI, will become a standing agenda item at future meetings.
- The APS AI Plan 2025 was cited as a reference point for AI's role in improving government efficiency and service delivery.
- AI governance is a minor thread in a broader digital identity, data sharing, and cyber security agenda.
-
New project aims to accelerate the safe adoption of autonomous shipping
- The Alan Turing Institute is launching a project to accelerate safe adoption of autonomous shipping technology.
- The project links AI governance and safety assurance to a specific high-stakes physical domain — maritime transport.
- Extracted text is truncated; full scope, methods, and outputs are unknown from this item alone.
-
Blog
- The Alan Turing Institute is hosting a public lecture on frontier AI resilience in April 2026.
- Limited extracted content - only an event listing with minimal detail on substance or outputs.
- Low signal for APS readers; a UK public lecture with no evident Australian policy connection.
-
Personalising healthcare with connected digital twins
- Alan Turing Institute blog explores using digital twins to personalise care for pulmonary arterial hypertension patients.
- Digital twins in healthcare raise governance questions around data integrity, model validity, and clinical decision support.
- Extracted text is sparse - substantive detail is limited; limited direct relevance to Australian federal AI governance.
-
Thu 12 Mar 2026 Tickets on sale now - 2026 Australian Government Data Forum Government Australian Government data
- The 2026 Australian Government Data Forum is scheduled for 19 March 2026, online and in Canberra.
- The forum covers data sharing, privacy, ethics, and cross-government collaboration - not AI specifically.
- Limited direct AI relevance; this is a data profession event with only adjacent AI implications.
-
Building the Strategic Supply Chain Network
- NIST workshop focuses on Open Knowledge Networks for US supply chain resilience and real-time visibility.
- AI and algorithmic systems are not the subject; data infrastructure and supply chain coordination are.
- Limited direct relevance to Australian federal AI governance work - included for completeness.
-
Professor Rebecca Eynon elected to prestigious Academy of Social Sciences Fellowship
- Oxford Internet Institute professor Rebecca Eynon elected to the Academy of Social Sciences Fellowship.
- Her research examines inequities arising from AI and digital technology use in education settings.
- Low direct signal for APS readers; this is an academic honour announcement, not policy or guidance.
- Week of 2 March 2026
-
Import AI 447: The AGI economy; testing AIs with generated games; and agent ecologies
- MIT/WashU/UCLA paper argues human 'verification bandwidth' becomes the binding constraint in an AI-driven economy.
- Biosecurity research finds LLMs provide 4.16x accuracy uplift to novices on bioweapon-related tasks - a concrete dual-use risk signal.
- Policy framing is speculative and US-centric; limited direct applicability to current Australian regulatory settings.
-
Can we create a clear understanding of what agentic AI is and does?
- OECD is working to establish a clear definitional framework for agentic AI and AI agents.
- Agentic AI is increasingly cited in Australian governance discussions - OECD framing may influence local definitions.
- Extracted text is a preview only; full content and analytical depth cannot be assessed from this excerpt.
-
Wed 04 Mar 2026 Seeking feedback on proposed changes to the Digital ID Amendment (Redress Framework) Rules 2025 Business, Government, Individuals Digital ID
- Finance is consulting on redress rules strengthening consumer protections within the Australian Government Digital ID System.
- Proposed changes include financial compensation powers and mandatory notification obligations for fraud and cyber incidents.
- This is a Digital ID governance item with minimal direct AI content - low signal for AI-focused APS readers.
- Week of 23 February 2026
-
DTA signs new 5-year agreement with Microsoft: Delivering value and innovation for the Australian Government
- DTA has signed a new five-year Volume Sourcing Agreement with Microsoft, commencing 1 July 2026.
- The agreement explicitly accelerates APS adoption of AI and other emerging technologies under whole-of-government terms.
- Enhanced legal provisions cover governance, security, liability, and handling of government data across all agencies.
-
Import AI 446: Nuclear LLMs; China's big AI benchmark; measurement and AI policy
- AI researcher Jacob Steinhardt argues investing in AI measurement tools is the highest-leverage AI governance intervention available.
- LLMs tested in nuclear wargame simulations escalated faster and more aggressively than humans, with near-universal tactical nuclear use.
- Chinese researchers have released ForesightSafety Bench, a large-scale AI safety evaluation framework largely mirroring Western safety benchmarks.
-
Deadline extension 20 March: Global call for ‘Governing with Artificial Intelligence’: Share your initiatives and insights on AI-driven innovation in government
- OECD invites governments to submit AI use cases, policy initiatives, and implementation tools for a global collection.
- Australian agencies could contribute examples of responsible AI use in government to this international knowledge-sharing exercise.
- Submission deadline is 20 March 2026; extracted content is minimal and detail on scope or outputs is not available.
-
Generating Harms: Generative AI’s Impact and Paths Forward
- EPIC's 2023 paper, spotlighted by MIT AI Risk Repository, maps nine harm categories from generative AI adoption.
- The taxonomy covers physical, economic, psychological, discriminatory, and dignitary harms with documented real-world examples.
- This is a secondary spotlight of an existing 2023 advocacy paper; not new primary research or Australian-specific guidance.
- Week of 16 February 2026
-
Announcing the "AI Agent Standards Initiative" for Interoperable and Secure Innovation
- NIST's CAISI launches an AI Agent Standards Initiative covering interoperability, security, and identity for autonomous AI agents.
- Three pillars target industry-led standards, open-source protocol development, and AI agent security research.
- Active RFIs on AI agent security (due 9 March) and agent identity/authorisation (due 2 April) are open to stakeholders.
-
New Report: Expanding the AI Evaluation Toolbox with Statistical Models
- NIST CAISI publishes AI 800-3, introducing statistical rigour improvements for AI benchmark evaluations.
- The report distinguishes benchmark accuracy from generalised accuracy - a gap that affects how agencies interpret AI procurement claims.
- Applies generalised linear mixed models to 22 frontier LLMs on three major benchmarks, demonstrating measurable uncertainty quantification gains.
-
The OECD’s new responsible AI guidance: A compass for businesses in a complex terrain
- OECD has published Due Diligence Guidance for Responsible AI to help businesses manage AI risks.
- Guidance aims to align business AI practices with global standards and support trustworthy AI value chains.
- Only a brief blog summary is available - full substance and APS applicability cannot be assessed from this excerpt.
-
Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment
- A 2023 paper catalogues LLM trustworthiness across 7 categories and 29 subcategories, now spotlighted by MIT's AI Risk Repository.
- The taxonomy covers reliability, safety, fairness, misuse resistance, explainability, social norms, and robustness — directly mapping to APS AI governance concerns.
- This is a secondary blog summary of a 2023 arXiv paper; the taxonomy itself is not new or Australia-specific.
-
Artificial Intelligence Trust, Risk and Security Management (AI TRiSM)
- MIT AI Risk Repository spotlights the AI TRiSM framework covering trust, risk, and security management across the AI lifecycle.
- Framework organises AI risks into bias/privacy, societal manipulation/deepfakes/LAWS, and security threat categories.
- This is a blog summary of a 2024 academic paper - useful reference material rather than actionable policy guidance.
-
Australia rises to second globally in the OECD Digital Government Index
- Australia ranked 2nd globally in the 2025 OECD Digital Government Index, scoring 88% across 42 countries.
- The APS AI Plan and responsible AI policy are cited as contributors to Australia's improved 'proactiveness' dimension ranking.
- AI governance is one thread in a broader digital government story; not the primary focus of this item.
-
CAISI to Host Listening Sessions on Barriers to AI Adoption
- NIST's CAISI is hosting virtual workshops in May 2026 on AI adoption barriers in healthcare, finance, and education.
- Outputs will inform CAISI's work on AI adoption confidence under the US AI Action Plan - potentially yielding reusable sector frameworks.
- Limited direct relevance to APS; findings may offer comparative insight for agencies with similar sector-specific AI deployment challenges.
-
Introducing the 2025 MSc Thesis Prize Winners
- Oxford OII awarded 2025 MSc thesis prizes including work on LLM evaluation validity and AI moderation legitimacy.
- The LLM benchmark validity thesis is directly relevant to APS work on AI procurement and capability assessment.
- An academic prize announcement - useful for identifying emerging research, not for immediate APS action.
-
After Orthogonality: Virtue-Ethical Agency and AI Alignment
- Essay argues AI alignment should be grounded in virtue ethics and practices rather than goal-directed reasoning.
- Proposes that AI agents sharing a 'practices-based' deliberative logic with humans would better support human agency.
- Philosophical alignment research - interesting framing but no immediate policy or operational implications for APS agencies.
-
Import AI 445: Timing superintelligence; AIs solve frontier math proofs; a new ML research benchmark
- Nick Bostrom argues superintelligence development should proceed now despite existential risk, framing delay as equally dangerous.
- Meta's Kunlun recommender system establishes scaling laws for recommendation AI, improving ad-targeting predictability at massive scale.
- Content is a research/opinion digest with limited direct APS policy or governance actionability at this time.