Model Routing
Reporting, analysis and primary-source coverage on Model Routing.

Agents & Tools
Model Context Protocol Transition to Agentic AI Foundation Formalises Enterprise Tool Calling Standard
The Model Context Protocol has transitioned to the Linux Foundation Agentic AI Foundation, establishing a vendor-neutral protocol for AI agent tool calling.
Clara Tan

Zero-Days & Patching
Critical GPUStack RCE Vulnerability CVE-2026-102895 Exposes Enterprise AI Compute Clusters
A critical unauthenticated remote code execution flaw in open-source orchestrator GPUStack allows threat actors to seize enterprise AI accelerator cluster…
Elena Rostova
Chips & Compute
Nvidia Approaches TSMC for Custom 3D DRAM Integration on Next-Generation AI Silicon
Nvidia has initiated technical talks with TSMC to integrate custom 3D DRAM directly onto logic tiles, targeting the memory bandwidth wall in AI clusters.
Marcus Vance

Models & Research
Google DeepMind AlphaProof Code Formally Verifies Signal Protocol Machine Code at AArch64 Level
DeepMind has used AlphaProof Code to construct end-to-end mathematical proofs verifying the Signal protocol machine code directly at the ARM64 assembly le…
Clara Tan

Agents & Tools
Vector Search and Retrieval at Scale: pgvector vs Pinecone vs Qdrant in Production (2026 Latency and Cost Analysis)
An in-depth evaluation of memory consumption, index build speeds, filtering latency, and operational cost across embedded PostgreSQL and dedicated vector …
Clara Tan

Funding & Deals
Seed and Series A AI Valuations Face Scrutiny as VCs Demand Repeatable Net Revenue Retention
Early-stage venture firms enforce stricter unit economics on application-layer AI startups, shifting capital into vertical enterprise workflows and develo…
Jin

Fintech & Commerce
Monetary Authority of Singapore Unveils SAFR Framework for Autonomous Agentic Financial Execution
Ahead of the Singapore FinTech Festival 2026, MAS establishes runtime security and liability boundaries for autonomous AI agents executing financial trans…
Rachel Lim

AI Safety & Policy
Coalition Sues OpenAI, Anthropic, Google, and Meta in San Francisco Court to Freeze Frontier Model Training
Digital safety watchdogs file an unfair competition suit in San Francisco Superior Court demanding an injunction against training next-generation frontier…
Clara Tan

Chips & Compute
Microsoft Lets Legacy Oracle GPU Deal Lapse as Cloud Spending Splits Between Maia 300 and Nvidia Rubin
Microsoft is allowing its 2023 Oracle AI capacity lease to expire because older Nvidia GPUs no longer fit agentic inference needs, underscoring a 3-year h…
Marcus Vance

Agents & Tools
AWS Says AI Inference Now Drives Over Half of AI Revenue as Amazon Pushes Trainium and Nova Ahead of re:Invent
With AWS running near a $182 billion annualised pace, leadership says inference now exceeds training revenue while capacity constraints fuel external Trai…
Clara Tan

Enterprise & Cloud
AWS Cuts Bedrock AgentCore Cold Starts 15x With Firecracker Snapshots and Ships Native Web Search API
Amazon Web Services made its V2 AgentCore runtime the default on 12 November, cutting 2 GB container cold starts from 29.7 seconds to under 2 seconds alon…
Aarav Mehta

Models & Research
OpenAI Splits GPT-6 API Into Three Pricing Tiers and Lists ChatGPT Enterprise on AWS Marketplace
OpenAI's November 2026 API price sheet formalises GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna with 90% prompt caching discounts while opening ChatGPT Enterpr…
Clara Tan