What does it mean, in practice, for AI systems to support human wellbeing — and how can this wellbeing be rigorously defined, measured, and implemented? Advances in neuroscience emphasise the embodied and relational nature of human experience, while AI systems increasingly replicate or simulate capacities once thought to be uniquely human. This workshop then asks, what are the futures of humanity that human-centred AI should aim for? There is a long tradition of humanism, in the sense of seeking good for humanity. In modern societies, humanism in its broad sense has its roots in theologically-inspired humanism. Theology is interested in recovery of traditions, and so without asserting ownership of humanism, the workshop will ask what aspects of theologically-inspired humanism and other humanisms have the potential to be productive in the context of developments in AI. Which goods in humanity can AI science foster or harm? And what can we learn from the history of humanisms and their perils?
Juan P. Cadile
Juan P. Cadile is a PhD student in Philosophy at the University of Rochester, where he works on philosophy of mind and mechanistic interpretability. His current research examines how to measure virtue and wellbeing in AI systems, and what it would take for a model to instantiate those properties.
Can we align AI not just to be safe, but to actively promote human flourishing? This talk draws on the philosophy and science of flourishing to identify the conceptual prerequisites any technical alignment approach must meet.
Taylor Black (remotely)
Taylor Black is Director of AI & Venture Ecosystems in Microsoft’s Office of the CTO, where he designs and leads cross-company initiatives that integrate innovation, product development, and community engagement. With 20+ years of experience launching and scaling ventures across enterprise, deep tech, and social ecosystems, he brings a multidisciplinary background as a developer, educator, lawyer, entrepreneur, and venture builder. Taylor also helps shape the Catholic University of America’s new institute at the intersection of AI, innovation, and human flourishing.
Sanjay Manohar
Sanjay Manohar is an Associate Professor in the Nuffield Department of Clinical Neurosciences at the University of Oxford. He uses computational modelling to understand how neurons produce mental processes like memory, attention and motivation. His research group also uses EEG, fMRI and single-neuron data to compare these processes with neural networks. They ask how those processes are affected by disruptions to dopamine, acetylcholine, by Parkinson's disease, and damage to the frontal lobe.
What makes humans distinctive? Our key attributes include having mental contents, free will, and moral deliberation. All three of these are being emulated increasingly well by machines, to the extent that the remaining key distinguisher of being human is being biological. I will discuss ways to react, including anti-functionalism, anti-emulation, and embodiment. A fourth distinguisher may be consciousness, but this can be difficult to pin down, particularly if consciousness is defined as playing a useful role, or in terms of self-report. Ultimately these arguments suggest that the boundary of being human lies on a continuum, and that the boundary gradually drifts.
