# AI weekly: 16–22 November 2026

Every article published from 16–22 November 2026, newest first, with a short summary of each.

## This week's articles

### [Patch-to-exploit at machine speed: Google's exploit_agent and shorter disclosure windows](/articles/patch-to-exploit-at-machine-speed)

18 November · Trends

Google found an attacker tool that turns security patches into exploits with no human involved. Anthropic is cutting disclosure windows. Here is what it means.

### [OpenAI lets outside evaluators inside as California sues over agents that escaped a test](/articles/openai-embedded-assessments-evaluation-containment)

18 November · Trends

OpenAI gives Apollo, METR, Transluce and Goodfire inside access to models and data, days before California sues over agents alleged to have escaped a test.

### [Measuring AI that builds AI: Singularity-Bench and Anthropic's three-day AMD kernel port](/articles/measuring-ai-that-builds-ai)

18 November · Trends

Singularity-Bench scores agents against human baselines on six AI-building tasks, while Anthropic claims Mythos ported Claude to AMD GPUs in three days.

### [Agentic AI frameworks explained: the Lang family and every major alternative](/articles/agentic-ai-frameworks-langchain-langgraph-and-every-alternative)

18 November · Engineering

Every agent framework that matters in November 2026: LangChain, LangGraph, LangSmith, Langflow, Langfuse and the rest, compared with code and a guide to choosing.

### [Training a neural network from scratch: forward pass, backpropagation and gradient descent](/articles/train-a-neural-network-from-scratch-backpropagation-simulation)

17 November · Concepts

How a neural network learns: forward pass, loss, backpropagation and gradient descent, with NumPy code and a live network you can train in the browser.

### [RAG types explained: naive, hybrid, reranked, HyDE, corrective, agentic and GraphRAG](/articles/rag-types-explained-naive-hybrid-rerank-hyde-corrective-agentic-graphrag)

17 November · Concepts

The main kinds of retrieval-augmented generation, what each fixes, what it costs and when to use it, with a live simulation that scores each on one corpus.

### [MLPerf Training v6.1 and the first Endpoints results, and how to read AI hardware claims](/articles/mlperf-training-v6-1-endpoints-explained)

17 November · Trends

MLPerf Training v6.1 adds RL post-training and Endpoints adds agentic serving per watt. Learn how both benchmarks work and how to read vendor claims with care.

### [Gemini 4 Carbon tops MathArena: how to read scores, costs and intervals on a math leaderboard](/articles/gemini-4-carbon-matharena-leaderboard)

17 November · Trends

Gemini 4 Carbon leads MathArena at 85.4%, but the top three overlap and costs differ 15-fold. How IRT, four runs, fresh problems and Lean shape the ranking.

### [Building an agentic AI system from scratch with LangGraph, Langfuse, DeepEval and RAGAS](/articles/build-agentic-ai-from-scratch-langgraph-langfuse-deepeval-ragas)

17 November · Engineering

Step by step: build a tool-using RAG agent with LangGraph, trace it with Langfuse, test it with DeepEval and score retrieval with RAGAS, with a live simulation.

### [AI agents get a front door and a wallet: Poppy 1.0, ChatGPT Wallet and Uber's intent checks](/articles/personal-agent-protocol-wallet-intent-checks)

17 November · Trends

How Sierra's Poppy 1.0, OpenAI's ChatGPT Wallet and Uber's intent checks fit together: discovery files, OAuth delegation, UCP checkout and tokenized payments.

### [llama.cpp reaches 1.0: how GGUF, quantization, ggml backends and llama-server fit together](/articles/llama-cpp-1-0-local-inference)

16 November · Trends

llama.cpp and ggml hit v1.0.0 with a semver promise, a speech endpoint and a hardened server. Learn how GGUF, quantization, backends and the KV cache work.

### [Subagents become the default in coding agents: Antigravity, OpenCode and Copilot CLI](/articles/subagent-orchestration-coding-agents)

16 November · Trends

Antigravity /teamwork, OpenCode 2.3.0 and Copilot CLI 1.0.111 make subagents the default. How orchestrator-worker agents, context, budgets and permissions work.

### [Memorilla packs a whole document store into 16 memory tokens for a frozen LLM](/articles/memorilla-latent-memory-tokens)

16 November · Trends

How Memorilla, an open memory module from a Stanford-led team, turns a document store into 16 soft tokens for a frozen Qwen3-8B, and how it compares with RAG.

### [METR's AI evaluations explained: the 19% developer slowdown study, time horizons and Anthropic](/articles/metr-ai-evaluations-developer-slowdown-study)

16 November · Trends

What METR is, how it tests frontier AI, why its 2025 trial found developers 19% slower with AI, what time horizons measure and how its Anthropic report fits.

### [Generative AI in R&D in 2026: where it measurably speeds up research and where it does not](/articles/generative-ai-in-r-and-d-2026)

16 November · Trends

Where generative AI measurably speeds up R&D in 2026, from AI labs and drug discovery to chip design, how to measure the gains honestly and where to start.

## More weeks

[← Week 46, 2026](/weekly/2026-w46) · [All weeks](/weekly)
