Michael Anslow

Michael Anslow is a 0 to 1 technology innovator, grounding new technological experiences in research and understanding them in the context of practical systems rather than in isolated academic contexts. His main technical background is in natural language processing, but more recently he has branched out into multi-modal modelling, particularly in the audio-visual domain. 

Michael’s early research addressed artificial language learning in the form of the iterative alignment of ontologies, but since then he has focused more on natural language processing in different contexts such as data analysis, data exploration and generative modelling. In recent years he explores creative assistance using deep learning as a primary technology. He created a writing assistant using masked language modelling (prior to ChatGPT) that was used by various musicians at Sony CSL – Paris in collaboration with the Music Team. Since then, he has moved on to multi-modal domains, initially text and audio and audio-visual experiences. He is particularly interested in using AI to create ambiances dynamically to support creative experiences. 

Alongside his technical work, Michael has also become very interested in the big questions of AI and society. To this end he co-organizes an interdisciplinary discussion group on AI and philosophy between Sony CSL – Paris and local researchers.