Most of my work at present is in the social impacts and governance of AI. I have worked on those topics both in New Zealand and internationally.

• In New Zealand, I have worked on impacts of AI on jobs, on AI in healthcare, and on uses of AI in the government. This latter work was influential in establishment of New Zealand's Algorithm Charter. 
• Internationally, I co-led the Global Partnership on AI's project on Social Media Governance. (The Partnership is now part of the OECD.) We  worked on several topics: in particular, detection mechanisms for AI-generated content (see here and here), governance of recommender algorithms (see here and here), and content moderation methods (see here). This work had influence on two pieces of EU legislation: the AI Act, and the Digital Services Act. I also participated in the Christchurch Call, and the Global Internet Forum to Counter Terrorism.  I'm a member of the OECD's network of AI experts (OECD.AI). 
• Recently I helped to found the Social Data Science Alliance, to help facilitate work conducted by external researchers inside large online platforms, under provisions created by the EU's Digital Services Act. 

My Cognitive Science work focusses on building computational models of the brain. I am particularly interested in language, and its relationship to the sensorimotor system - how can we talk about the things we see, and the things we do? I wrote a book, ‘Sensorimotor Cognition and Natural Language Syntax’ (MIT Press, 2012) that offers an answer to this question; I have given many talks about this work, including TEDx talks in Auckland and Athens. I also have expertise in computational models of discourse and dialogue structure, and have published widely in these areas too. Latterly, some of this work was pursued in collaboration with the AI company Soul Machines: I worked with their research group, to help design a computational brain model, and I also advised the company about its commercial product, a human-computer dialogue system.