The Government Order That Sent Anthropic's Newest AI Models Dark
Three days after Fable 5 reached the public, Anthropic said a U.S. export-control directive forced it and Mythos 5 offline worldwide for every customer.
Technology intelligence
AI Infrastructure Insights
AI compute, model deployment, agent infrastructure, and the technical stack behind enterprise AI adoption.
32 insights · Updated weekly · Editorially independent
Choose a lens
Filter by topic without losing your placeLatest stories
10 items in AI Infrastructure
Singapore Became Asia's AI Launchpad. It May Never Be the Lab.
Singapore has made itself the place the AI industry routes through — capital, founders, regulators, scale. What it hasn't built is the frontier research that would make it indispensable. The ceiling is real, and largely self-imposed.
Multi-Sensory AI: Models That See, Hear and Touch Reach the Enterprise
Multimodal AI is moving past text and images to ingest sensor, thermal and tactile data in real time. As models gain sub-200ms perception, physical AI is reaching factories, ports and logistics — and reshaping edge-infrastructure demand.
Andrej Karpathy Joins Anthropic's Pre-Training Team, Pauses Eureka Labs
Andrej Karpathy joined Anthropic's pre-training team on 19 May 2026 — his first full-time lab role since leaving OpenAI in Feb 2024 — to stand up a team using Claude to accelerate Claude's own training.
AI Engineer Lands in Asia: Singapore Edition Draws 2,000 Builders, Cabinet-Level Keynotes
AI Engineer Singapore — the first Asia edition of the AI Engineer Summit series — drew about 2,000 in-person attendees and 50,000 online viewers to The Capitol Theatre from 15–17 May 2026.
Anthropic's MCP Becomes a De Facto Integration Standard as Claude 4.x Drives Enterprise Pull
The Model Context Protocol (MCP), introduced by Anthropic in November 2024, has — over the course of 2025 — quietly become the integration layer of choice for connecting LLMs to…
OpenAI's Agent Era: From ChatGPT to Operator, Deep Research, and the Responses API
After spending much of 2024 hardening foundation models, OpenAI's late-2025 product cadence pivoted decisively toward autonomous agents. Operator — the company's browser-based a…
Andrej Karpathy's Eureka Labs and the Open-Source Curriculum for AI-Native Engineers
Since departing OpenAI in February 2024, Andrej Karpathy has channelled his attention into Eureka Labs — an AI-native education company aimed at rebuilding how technical subject…
Claude Code: Anthropic's AI Coding Agent Is Reshaping Developer Workflows
Anthropic launched Claude Code — a terminal-native AI coding agent — as a generally available product in early 2025, and adoption across enterprise software teams has accelerate…
OpenAI's o3 and GPT-5 Signal a New Era of Enterprise AI Deployment
OpenAI's release of o3 — and the subsequent preview of GPT-5 — has significantly shifted the competitive landscape for enterprise AI. The o3 model, designed for advanced reasoni…