Latest
OpenAI has added full-screen UI, conversation panels, file editing and MCP Events to ChatGPT plugins. We break down what they can do and where the limits lie across four layers: shared standards, ChatGPT-specific extensions, event subscriptions, and distribution and permissions.
We compare OpenAI's GPT-6.1 Sol with the previous Sol, Astra, and Claude, using tables and charts to show performance and per-task cost by setting, plus a cache-write pricing estimate.
OpenAI has announced dots, an always-on agent that works on its own cloud computer. Here is a look at eligible plans and usage limits, which actions need approval, the permission mix-ups that remained in OpenAI's published evaluation, and Meta's business-focused Muse expansion.
Reuters reports on Anthropic's IPO prospectus, separating the roughly $42 billion net loss from the $518 billion in future payment obligations, while an Akamai contract disclosure reveals the terms, cancellation clauses, and expansion options behind such long-term deals.
The Wall Street Journal reports OpenAI halted the release of GPT-6.1 Astra. Primary-source data on the current Astra model reveals how average safety improvements and serious scope violations can coexist.
A University of Vienna research team activated a photonic quantum processor aboard a small satellite, successfully measuring two-photon interference and unitary transformations in orbit. Despite losing half its detectors and suffering laser degradation, the experiment demonstrated the resilience of photonic quantum operations in the space environment.
Instinct, an AI agent still in early access, has raised $1 billion. We examine the transaction volume, compute resources, and delegated authority behind its $10 billion valuation.
The National University of Singapore reports the smallest systematic uncertainty yet achieved with a lutetium optical atomic clock. Here's what a 200-hour comparison of two clocks showed—and what still stands between this result and a redefinition of the second.
BlueQubit reports generating 1 million samples in 19 seconds using 61 qubits on IBM's Nighthawk r2. We examine the roughly 110-year classical-computing estimate and how the output quality was verified, based on the preprint paper.
A draft on the UK government's AI Knowledge Hub advises staff to skip thank-you-only messages, choose the smallest model that can do the job, and keep prompts short. Set against existing policy, it also shows how hard the savings are to measure.
