You got AI in my Neuroscience!

Bernstein 2024 presentation

30th September, 2024

In 2017, Neuromorphic engineering was in a similar state as ANN research in 2004. Moving GPUs into mainstream compute required a confluence of three things: hardware optimised for a basic computational element, a useful programming model, and the API and tooling to make deployment easy. In this talk I argue that Neuromorphic processors can follow the same path, with gradient descent as the programming model and simple integrate-and-fire neurons as the computational element, drawing on our experience commercialising SNN hardware at SynSense. I also discuss what the academic community can do to support widespread adoption of Neuromorphic compute. I presented this talk remotely at the Bernstein Conference 2024.

Governing autonomous systems at machine speed

An interview on AI governance and LexChip

5th August, 2026

Humans are not capable of practically overseeing the actions of autonomous systems, as these accelerate and proliferate. In this interview I talk about governing autonomous systems by their behaviour, in the same way the law governs humans, and how LexChip uses consent-based smart legal contracts to bring real-time oversight, breach detection and a legally admissible record to autonomous AI deployments.

Spikes, wiring, and general principles

An interview on neuroscience and spiking neural networks

29th July, 2026

Is spike timing essential to cortical computation, or just an efficient way to transmit information? In this interview I discuss moving from modelling cortical computation in academia to engineering spiking neural networks at SynSense; the search for general principles in neocortex; and what a complete connectome would and would not tell us.