AI resources library
Contents:
1. Primers and foundational guides
- GenAI-KM organizational readiness self-assessment tool
- GenAI Concepts
- The AI Tech Stack
- ISO/IEC 42001:2023 guide for business
- MITRE AI Maturity Model
- The Geopolitics of AI: Decoding the New Global Operating System
- Evaluating an AI-generated podcast
2. AI Agents
3. Ethical and responsible AI, AI governance
- PwC’s 2025 Responsible AI survey: From policy to practice
- AI Ethics and Governance in Practice Program
- Architectures of Global AI Governance: From Technological Change to Human Choice
- Global AI Law and Policy Tracker
- UNESCO Global AI Ethics and Governance Observatory
- UN General Assembly appoints Independent International Scientific Panel on Artificial Intelligence
4. Risk management and safety
- Guidance for Risk Management of Artificial Intelligence systems
- AI Risk Mitigation Taxonomy
- International AI Safety Report 2026
- AI Incident Database tracks reports of AI harms
1. Primers and foundational guides
GenAI-KM organizational readiness self-assessment tool
To help organizations to have a sound basis for their deployment of generative AI (GenAI) in knowledge management (KM), the GenAI-KM organizational readiness self-assessment tool works across five strategic dimensions.
The tool converts each strategic dimension and sub-dimension into targeted assessment questions that managers can use to evaluate their organization’s readiness for GenAI adoption in KM processes. By investigating each question, priority improvement areas can be identified before or during implementation.
The tool is one of the outputs of a bibliometric literature review analysing 1411 articles related to the topic of Gen AI in KM.
GenAI Concepts
To help entities interested in GenAI deployments, the GenAI Concepts publication outlines 42 concepts fundamental to AI software systems. Each concept is illustrated through descriptions, examples and real-world use cases, with accessible language and visual elements to accommodate a diverse range of stakeholders and readerships.
GenAI Concepts was developed as a collaboration between two Australian organisations, the ARC Centre of Excellence for Automated Decision-Making and Society(ADM+S) and the Office of the Victorian Information Commissioner(OVIC). It was produced through expert consultation and analysis of academic literature, industry reports, and policy guidance.
With thanks to Peter Slattery, PhD on LinkedIn.
The AI Tech Stack
The AI Tech Stack report serves as a comprehensive primer, offering public policy and cybersecurity practitioners insights into this dynamic landscape where their domains increasingly intersect.
The AI technology stack comprises five distinct yet interdependent layers:
- Governance layer – The framework that effectively wraps around the whole AI Technology Stack—a layer that aims to ensure responsible deployment through security protocols, legal constraints, ethical principles, and policies.
- Application layer – The user interface that transforms complex AI capabilities into accessible tools through browsers, APIs, dashboards, and other user interfaces.
- Infrastructure layer – The essential computational foundation that powers AI systems, enabling the intensive demands of training and inference through specialized hardware, cloud platforms, and energy resources.
- Model layer – The core computational component that processes data according to sophisticated algorithms to recognize patterns and generate predictions or decisions. This includes the machine learning approaches that enable systems to learn without explicit programming.
- Data layer – The foundation of AI systems, providing the raw material that fuels models. The quality, diversity, and quantity of this data largely determine the intelligence and capabilities of the final model.
Robust security across this stack is a technical necessity and a strategic imperative. AI security extends traditional cybersecurity concepts to confront unique vulnerabilities within machine learning systems, including adversarial attacks, model poisoning, and data exploitation. Organizations that prioritize comprehensive AI security not only mitigate risks but also position themselves as leaders in tomorrow’s innovation networks, capable of rapidly integrating advancements while sustaining trust. By embedding security measures early in the development process, organizations gain downstream competitive advantages, including faster deployment cycles, greater stakeholder confidence, and better products. The first step to this process is understanding the AI Tech Stack.
With thanks to Peter Slattery, PhD on LinkedIn.
ISO/IEC 42001:2023 guide for business
Also supporting ethical and responsible AI in KM for business, the free report Understanding 42001: Guide for Australian Business has been published by Standards Australia, CSIRO, and Australia’s National Artificial Intelligence Centre. Although developed in an Australian context, much of the content is more widely applicable.
The report advises that ISO/IEC 42001:2023 addresses the need for consistency and ethical implementation of AI across borders. There is a knowledge gap across organisations when it comes to managing the potential risks and complexities of AI. Data gathered for the latest Australian Responsible AI Index report found that although the vast majority (82%) of companies surveyed believed they were taking a best-practice approach to responsibly using AI in their businesses, less than a quarter (24%) had any measures in place to ensure that was what they were actually doing. These figures are mirrored globally.
The report helps to address this knowledge gap by providing information on ISO/IEC 42001, its benefits, and how it will impact organisations and the broader community. It includes information on the purpose of ISO/IEC 42001, its development, and the importance of certification.
MITRE AI Maturity Model
The MITRE Artificial Intelligence (AI) Maturity Model (MM) can be viewed as a methodology to provide guidance and recommendations for building a foundation for successful AI implementation across an organization. It could potentially be integrated with KM maturity models.
It was developed based on a systematic review of commercial AI MMs extant throughout the private sector as well as an assessment of both the Capability Maturity Model Integration (CMMI) appraisal processes developed by Carnegie Mellon University and the National Institute of Standards and Technology’s AI Standards.
The AI MM is organized according to six pillars that industry considers major aspects of maturity that are key to successful AI adoption:
- Ethical, equitable, and responsible use.
- Strategy and resources.
- Organization.
- Technology enablers.
- Data.
- Performance and application.
Each pillar has either three or four dimensions (20 total) describing specific actions and activities that demonstrate advancing mastery of AI maturity for that dimension. These pillars and dimensions are assessed across five readiness levels that qualitatively describe different approaches to AI adoption. They are juxtaposed with five assessment levels intended to describe hierarchical and scalable progress throughout AI adoption: Initial, Adopted, Defined, Managed, and Optimized.
With thanks to Peter Slattery, PhD on LinkedIn.
The Geopolitics of AI: Decoding the New Global Operating System
A report from the JPMorganChase Center for Geopolitics provides what appears to be the first comprehensive insight into the geopolitics of AI, with AI advancing in very different directions across the world.
Key takeaways from the report are:
- China and the U.S. dominate, but on divergent paths. Beijing pursues state-led self-reliance and lower cost open-source exports, while Washington bets on private-sector innovation, infrastructure buildout, and defense integration. This is a key geopolitical fault line, as countries may face a choice which direction to go.
- Tech sovereignty and standards are fragmenting the field. More countries, and in some cases localities, are asserting control over AI infrastructure, talent, and governance frameworks—exporting rules as well as building walls, and forcing firms to navigate a more divided AI ecosystem.
- Energy and hardware are the new chokepoints. Semiconductors, critical minerals, and electricity capacity define who can scale AI, and who risks falling behind.
- Capital is repositioning the map. Middle Eastern sovereign wealth funds are leveraging energy abundance to become key players in AI infrastructure.
- AI is transforming defense and deterrence. From swarming drones to AI-enabled decision loops, militaries that integrate AI fastest will hold decisive battlefield advantages.
While AI is advancing in many directions, seven strategic “axes” stand out for their significance to the current geopolitical moment. Each reflects a distinct dimension of competition, cooperation, and national ambition—and each is already motivating governments, businesses, and alliances to reposition in ways that will shape the century ahead.
The report authors advise that, “To best navigate the years ahead, understanding these seven strategic axes is essential.”
Evaluating an AI-generated podcast
In a LinkedIn post, Dr Sarah Cummings reports on her experimental use of NotebookLM to see whether it is an effective tool for learning and advocacy. In her analysis, Sarah explores the positives and negatives of the experience, and puts forward useful lessons that others could learn from. Sarah used NotebookLM to generate a ‘Welcome to the Deep Dive’ podcast about her and Gerrit-Jan van Uffelen’s recent paper1 advocating for a knowledge agenda for food systems resilience in protracted crisis in the Horn of Africa.
Reference:
1. Cummings, S., & van Uffelen, G. J. (2025). A Knowledge Agenda for food systems resilience in protracted crisis in the Horn of Africa. Food Security, 1-16.]
2. AI Agents
The AI Agent Index
Published in February 2026, The 2025 AI Agent Index documents the origins, design, capabilities, ecosystem, and safety features of 30 prominent AI agents based on publicly available information and correspondence with developers. As shown in the following figure from the Index, 2025 has seen a sharp increase in interest in AI agents.
Key findings from the The 2025 AI Agent Index include:
- Rapid Deployment – 24 / 30 agents launched or received major agentic updates in 2024-2025, with releases accelerating sharply. Autonomy levels are rising in parallel.
- Autonomy Split – Chat agents maintain lower autonomy (Level 1-3), browser agents operate at Level 4-5 with limited intervention, and enterprise agents move from Level 1-2 in design to Level 3-5 when deployed.
- Transparency Gap – Of the 13 agents exhibiting frontier levels of autonomy, only 4 disclose any agentic safety evaluations. Developers share far more information about capabilities than safety practices.
- Foundation Model Concentration – Almost all agents depend on GPT, Claude, or Gemini model families, creating structural dependencies across the ecosystem.
- No Standards – There are no established standards for how agents should behave on the web. Some agents are explicitly designed to bypass anti-bot protections and mimic human browsing.
- Geographic Divergence – Agent development concentrates in the US (21/30) and China (5/30), with markedly different approaches to safety frameworks and compliance documentation.
With thanks to Peter Slattery, PhD on LinkedIn.
Advances and Challenges in Foundation Agents book
With more than 1,600 references, the academic book Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systems provides a comprehensive overview of AI agents. It frames intelligent agents within modular, brain-inspired architectures that integrate principles from cognitive science, neuroscience, and computational research. The exploration is structured into four interconnected parts:
- First, a systematic investigation of the modular foundation of intelligent agents, systematically mapping their cognitive, perceptual, and operational modules onto analogous human brain functionalities and elucidating core components such as memory, world modeling, reward processing, goal, and emotion.
- Second, a discussion of self-enhancement and adaptive evolution mechanisms, exploring how agents autonomously refine their capabilities, adapt to dynamic environments, and achieve continual learning through automated optimization paradigms.
- Third, an examination of collaborative and evolutionary multi-agent systems, investigating the collective intelligence emerging from agent interactions, cooperation, and societal structures, highlighting parallels to human social dynamics.
- Finally, addressing the critical imperative of building safe and beneficial AI systems, emphasizing intrinsic and extrinsic security threats, ethical alignment, robustness, and practical mitigation strategies necessary for trustworthy real-world deployment.
By synthesizing modular AI architectures with insights from different disciplines, this book identifies key research gaps, challenges, and opportunities, encouraging innovations that harmonize technological advancement with meaningful societal benefit.
3. Ethical and responsible AI, AI governance
PwC’s 2025 Responsible AI survey: From policy to practice
Articles in RealKM Magazine‘s artificial intelligence series have highlighted the need for frameworks for the responsible use of artificial intelligence (AI) in knowledge management (KM).
PwC’s 2025 Responsible AI survey: From policy to practice reveals that not only does responsible AI benefit the safe and trustworthy use of AI in KM, it is also becoming a driver of business value. Nearly 60% of executives say responsible AI boosts ROI and efficiency, and 55% report improvements in customer experience and innovation.
The focus is now changing to operationalization – turning Responsible AI principles into scalable, repeatable processes – with half of the survey respondents citing this as their biggest hurdle. The growing range of research and resources on ethical and responsible AI in RealKM Magazine‘s artificial intelligence series can greatly assist with this.
AI Ethics and Governance in Practice Program
The Turing Institute’s AI Ethics and Governance in Practice Program is composed of eight modules that provide end-to-end guidance to help AI project teams put ethical values and practical principles into practice across the AI project lifecycle.
Central to this guidance is the Process-Based Governance (PBG) Framework, a multi-tiered governance model designed to assist AI project teams in ensuring that the AI technologies they build, procure, or use are ethical, safe, and responsible.
The Program is designed for the public sector, but is likely to be more widely applicable, including to organizations that are carrying out AI consulting work for public sector agencies.
Architectures of Global AI Governance: From Technological Change to Human Choice
Articles in RealKM Magazine‘s artificial intelligence series have also highlighted the need for countries and international institutions such as United Nations (UN) agencies to have sound governance arrangements to address the impacts and risks of artificial intelligence (AI).
How can they govern this changing technology, in a rapidly changing world, using governance tools that may themselves be altered by AI? The open access book Architectures of Global AI Governance: From Technological Change to Human Choice provides conceptual and practical tools to tackle this question.
The book argues that, in crafting the global AI governance architecture, we must reckon with three facets of change:
- sociotechnical changes in AI systems’ impacts
- AI-driven disruptions to the fabric of international law
- political changes in the global AI regime complex.
Rather than just being an inquiry into how to govern AI, the book explores the changing face of global cooperation in the AI era, and how we can safeguard human choice over a future of transformative technological change.
Global AI Law and Policy Tracker
The IAPP has established the Global AI Law and Policy Tracker which identifies AI legislative and policy developments in a subset of countries across six continents (however, due to the speed and extent of policymaking, the tracker is not able to include all AI initiatives). The tracker also offers brief commentary on the broader AI context and related developments and identifies laws or policies in parallel professions like privacy.
Countries worldwide are designing and implementing AI governance legislation and policies in response to the speed and variety of proliferating AI-powered technologies. Efforts include the development of comprehensive legislation, focused legislation for specific use cases, national AI strategies or policies, and voluntary guidelines and standards.
There is no standard approach toward bringing AI under state regulation, however, common patterns toward reaching the goal of AI regulation can be observed. Given the transformative nature of AI technology, the challenge for jurisdictions is to find a balance between innovation and regulation of risks. Therefore, governance of AI often, if not always, begins with a jurisdiction rolling out a national strategy or ethics policy instead of legislating from the get-go. This pattern is evident throughout the tracker, and one such example was recently critiqued here in RealKM Magazine.
The IAPP is a policy neutral, not-for-profit association founded in 2000 with a mission to define, promote and improve the professions of privacy, AI governance and digital responsibility globally.
UNESCO Global AI Ethics and Governance Observatory
The aim of the UNESCO Global AI Ethics and Governance Observatory is to provide a global resource for policymakers, regulators, academics, the private sector and civil society to find solutions to the most pressing challenges posed by artificial intelligence.
The Observatory showcases information about the readiness of countries to adopt AI ethically and responsibly. It also hosts the AI Ethics and Governance Lab, which gathers contributions, impactful research, toolkits and good practices across a range of issues related to AI ethics, governance, responsible innovation, standards, institutional capacities, generative AI, and neurotechnologies.
UN General Assembly appoints Independent International Scientific Panel on Artificial Intelligence
At its 72nd Plenary meeting on 12 February 2026, the United Nations (UN) General Assembly appointed the 40 members of the new Independent International Scientific Panel on Artificial Intelligence.
In announcing the appointment, UN Secretary-General António Guterres states that:
We now have a multidisciplinary group of leading AI experts from across the globe, geographically diverse and gender-balanced, who will provide independent and impartial assessments of AI’s opportunities, risks and impacts – including to the new Global Dialogue on AI Governance.
In a world where AI is racing ahead, this Panel will provide what’s been missing – rigorous, independent scientific insight that enables all Member States, regardless of their technological capacity, to engage on an equal footing.
The creation of the Panel fulfills one of the commitments made in the Global Digital Compact, which was adopted at the 2024 Summit of the Future and is an annex to the Pact for the Future.
The 40 Panel members were selected from more than 2,600 candidates, after independent review by the International Telecommunication Union (ITU), the UN Office for Digital and Emerging Technologies, and UNESCO.
4. Risk management and safety
Guidance for Risk Management of Artificial Intelligence systems
The publication Guidance for Risk Management of Artificial Intelligence systems aims at providing valuable insights and practical recommendations to help identify and mitigate common technical risks associated with AI systems, helping in the protection of personal data.
The guide has been produced by the European Data Protection Supervisor for European Union Institutions, Bodies, Offices and Agencies (EUIs). However, it has wider application, particularly as the risk management approach put forward aligns with ISO 31000:2018 – Risk management — Guidelines.
With thanks to Peter Slattery, PhD on LinkedIn.
AI Risk Mitigation Taxonomy
The AI Risk Mitigation Taxonomy is an aspect of the MIT AI Risk Initiative, which aims to increase awareness and adoption of best practice AI risk management across the AI ecosystem. The MIT AI Risk Initiative has previously produced the AI Risk Repository.
The AI Risk Mitigation Taxonomy has three parts:
- The AI Risk Mitigation Database captures 831 mitigations extracted from 13 existing frameworks and classifications of AI risk mitigations.
- The Draft AI Risk Mitigation Taxonomy organizes mitigations into 4 main categories and 23 subcategories.
- The Taxonomy Categories include: Governance & Oversight Controls, Technical & Security Controls, Operational Process Controls, and Transparency & Accountability Controls.
With thanks to Peter Slattery, PhD on LinkedIn.
International AI Safety Report 2026
The newly published second edition of the International AI Safety Report builds on the mandate by world leaders at the 2023 AI Safety Summit to produce an evidence base to inform critical decisions about general-purpose artificial intelligence (AI).
Notable developments since the publication of the first International AI Safety Report include:
- General-purpose AI capabilities have continued to improve, especially in mathematics, coding, and autonomous operation.
- Improvements in general-purpose AI capabilities increasingly come from techniques applied after a model’s initial training.
- AI adoption has been rapid, though highly uneven across regions.
- Advances in AI’s scientific capabilities have heightened concerns about misuse in biological weapons development.
- More evidence has emerged of AI systems being used in real-world cyberattacks.
- Reliable pre-deployment safety testing has become harder to conduct.
- Industry commitments to safety governance have expanded.
With thanks to Peter Slattery, PhD on LinkedIn.
AI Incident Database tracks reports of AI harms
The AI Incident Database has been establish in response to the unforeseen and often dangerous failures that are occurring when artificial intelligence (AI) systems are deployed in the real world.
A repository of problems experienced in the real world can be an input into organizational frameworks for ethical and responsible AI in knowledge management (KM), as discussed in a previous RealKM Magazine article and in other resources on this page. In doing this, organizations can mitigate or avoid the risk of experiencing the same bad outcomes.
Header image source: Created by Bruce Boyes with Microsoft Designer Image Creator.
