When talking about local inference, Sam Altman says supply-constrained DRAM is better used in data centers than sitting in PCs and "being used 5% of the time" (Andrew E. Freedman/Tom's Hardware)

More in Artificial Intelligence

• As AI models commoditize, lasting wealth will accrue to those controlling chokepoints like distribution, payment rails, and a stablecoin dollar, not to the labs (The Commanding Heights)

  Economic power is shifting toward critical infrastructure as foundation models experience rapid commoditization. Controlling key leverage positions such as distribution channels, payment rails, and a stablecoin dollar will ultimately yield durable profits rather than the underlying development laboratories. This structural reality means financial gains generated by widespread technology diffusion will primarily benefit entities running vital access networks.

  1 min read
• Google sent TPUs to space, CEO Sundar Pichai shares a 'Hello World' update

  Project Suncatcher marks an experimental pivot to evaluate whether orbital data centers can eventually alleviate land and energy demands on Earth. By testing data center-class hardware in orbit, the initiative explores the viability of solar-powered computing outside terrestrial constraints. These initial communication passes establish that open-source models can execute operational tasks aboard orbiting satellites under real-world space conditions.

  2 min read
• Prospero: a new coding harness for a more civilized age

  Developed in C++, this lightweight harness allows artificial intelligence models to engage directly with development environments while using roughly two orders of magnitude less memory than standard web-stack frameworks. The low resource footprint specifically prevents operational crashes caused by system out-of-memory killers during intensive workloads such as software fuzzing campaigns. By requiring external connections to local engines or commercial APIs, it establishes a flexible, decoupled foundation for aut…

  9 min read