A chronological guide to my essays, talks, media commentary, and experiments on generative AI, evaluation, governance, and sociotechnical systems.
2026
Beyond Australia’s National AI Plan: Governing in the Gaps: A presentation I gave to a major Australian law firm on Australia’s fragmented AI governance landscape and the practical work needed to make accountability operational.
The Ghost in the Machine Speaks with an American Accent: Research on value drift in early frontier model, GPT3.
System Prompts in a Prompted Universe: An argument for treating system prompts as governance artefacts that shape behaviour without fully determining it.
AI Ethics roles in 2026: A short analysis of how AI ethics roles are changing as governance, evaluation, safety, and deployment accountability become harder to separate.
2025
Australia’s new National AI Plan: A response to Australia’s National AI Plan, focused on capability, risk, accountability, deepfakes, sustainability, and emerging agentic systems.
The Prompted Universe: Quantum Participatory Realism and Generative AI: An exploratory essay connecting participatory realism, prompting, and the way generative AI systems take shape through interaction.
Using Generative AI in Research: A practical reflection on how generative AI can support research while preserving scholarly judgement, accountability, and methodological care.
Vibe-Governing: Did AI (Mis)Calculate Trump’s Global Tariff Chaos?: A critique of AI-assisted political and economic sense-making when simulation, speculation, and governance judgement begin to blur.
2024
OpenAI partners with NewsCorp.: A response to the OpenAI and News Corp partnership, focused on data access, media power, and the political economy of generative AI.
How artificial intelligence could influence Australia’s next federal election: Media commentary on how AI-generated content, persuasion, deepfakes, and platform dynamics could affect Australian electoral politics.
EXPERT REACTION: Australia signs international AI declaration – what next?: Invited to provide xpert commentary on Australia’s participation in international AI governance and the practical work needed beyond symbolic commitments.
2023
Using GPTs to assist with AI governance policy.: A practical experiment in using custom GPTs to support AI governance policy work without outsourcing judgement or responsibility.
Lecture at Uni Sydney Business School: Slides and notes from a lecture on generative AI, ethics, and organisational responsibility.
Paws and perspectives: inclusivity in GenAI evaluation design.: A playful but serious reflection on inclusivity, perspective-taking, and the design assumptions built into generative AI evaluation. Presented at CSIRO conference.
AI Constitutions, Visual Subjectivity, and Experiments with GPT4-V(ision).: An experiment with GPT-4V that asks how visual interpretation, constitutional rules, and model subjectivity shape AI outputs.
Acknowledged for contributions to Australia’s eSafety Commissioner’s report.: A note recognising contributions to national work on generative AI, online safety, and regulatory response.
Generative AI is Human.: An argument that generative AI must be understood through the human systems, values, labour, and institutions that produce and use it.
ChatRegs23 An AI policy thinktank: Event materials from ChatRegs23, a policy-focused gathering on generative AI, regulation, and responsible governance.
Cave Painting and GenAI.: A reflection on generative AI, symbolic expression, and the long human history of making meaning through external representations.
I asked GPT4 to “Explain why language models cannot have sentience.”: A short experiment using GPT-4 to clarify why language modelling should not be confused with sentience or consciousness.
Your brain and predictive processing: An accessible explanation of predictive processing and why perception is better understood as active inference than passive reception.
X-riskers think differently: A critique of the assumptions, priorities, and reasoning styles that shape existential-risk approaches to AI.
Australia’s AI Acid Test: A response to Australia’s AI governance moment, asking whether policy settings can meet the practical risks of generative AI.
The reversed Turing Test.: A provocation about whether generative AI now tests human interpretation, projection, and judgement more than machine intelligence.
The Genie on Your Couch: A public-facing reflection on ChatGPT as a domestic, social, and psychological presence rather than merely a technical tool.
When the ‘Godfather of AI’ quits, we all need to start thinking: Media commentary on Geoffrey Hinton’s resignation and what it signalled for public debate about AI risk.
ChatGPT chief asks to be regulated, but does he mean it?: Media commentary on OpenAI, regulation, and the gap between public calls for oversight and enforceable accountability.
ChatLLM23: Event materials from ChatLLM23, a public conference on large language models, ethics, governance, and social consequence.
A conversation with Alex Hanna: Notes from a conversation with Alex Hanna on AI ethics, data, power, institutions, and critical approaches to machine learning.
2021
My group, PhD Students in AI Ethics, put on a conference.: A record of an early AI ethics conference organised by my PhD in AI Ethics group to widen debate beyond technical and institutional gatekeepers.
Chats with GPT-3: Memetics: An early experiment with GPT-3 exploring memetics, language generation, and the strange interpretive habits prompted by large language models.
2020
The Intelligence of Organisations: A reflection on organisational intelligence, distributed cognition, and why responsibility cannot be reduced to individual decision-making.
The Quiet Crisis of PhDs – Not so Quiet Anymore!: A critique of the structural pressures facing PhD students and the institutional conditions that make doctoral work precarious.
First questions: sociotechnial approaches to Ethical AI.: An early statement of the sociotechnical questions that should frame ethical AI beyond abstract principles or technical fixes.
How a toilet paper crisis can break AI.: A COVID-era example of how brittle AI systems can become when social behaviour shifts faster than their assumptions.
The quiet crisis of PhDs and COVID-19: A reflection on how COVID-19 intensified existing problems in doctoral education, labour, care, funding, and institutional support.
Citizens at the centre of AI ethics.: An argument for placing citizens, publics, and affected communities at the centre of AI ethics and governance.