Every Pixel is a Prediction

Generative AI, spatial computing and the question of trust

The image in front of you was not captured. It was inferred.

Upscaling, frame generation, neural rendering, Gaussian splatting and passthrough mixed reality have quietly changed what a display shows. More and more of what we see is a model’s best guess at what should be there, not a record of what is there.

This talk, delivered with a live Apple Vision Pro demonstration, takes the audience inside a headset where every pixel is reconstructed. It then asks the question that matters to boards, regulators and architects: when the visual world is generated, what can we still trust?

It is also an argument about accountability. A digital twin, a courtroom reconstruction or an operational dashboard is only as credible as its provenance. The organisations that win the AI-built world will be the ones that can prove where their pixels came from.