DiffusionGemma applies diffusion-style decoding to text, generating chunks in parallel and showing another path beyond autoregressive language models.
The important point is not just the headline figure. It shows how AI spending is moving into infrastructure, regulated workflows and distribution channels, where financing, compute access and policy exposure decide who can actually scale.
Sources: CocoLoop, Google AI Releases DiffusionGemma, a 26B MoE Open Model Using Text Diffusion for Up to 4x Faster Generation（MarkTechPost）；Google's DiffusionGemma generates 256 tokens in parallel and self-corrects as it goes（VentureBeat）；DiffusionGemma: 4x faster text generation（Google Blog）