The Token Reporter

A day in AI

Monday, 12 October 2026

40 stories · 2 from multiple sources

▲ 2 signals · 2 sources

AI models may be capable of self-replicating without access to a datacenter

Frontier Red Team

anthropic.com · 29 days ago

'AI' Is Not a Tool by Rich Hickey

gist.github.com · 29 days ago

openbmb/MiniCPM-V-4.7-35B-A3B

reddit.com · 29 days ago

More stories

Hello Python 3.15, goodbye 3.10, and now I can use 3.11

simonwillison.net · 29 days ago

AI Is a Tool. Using It Well Takes Skill.

yyhh.org · 29 days ago

datasette-ports 0.4

simonwillison.net · 29 days ago

Epic Games tells employees to use AI to get Disney game done

avclub.com · 29 days ago

Just say no to OpenAI and Anthropic

micahflee.com · 29 days ago

Show HN: Opus 5.5 made a fully featured Figma clone in 3 hours / 8 prompts

mana.space · 29 days ago

Also today

OCP Summit 2026: Powering AI with Open Innovations in Scale-Up, Scale-Out, and Scale-Across Networks

engineering.fb.com

Launch HN: Agent37 (YC F26) – Low-cost agent sandboxes for running agents

news.ycombinator.com

Reverse engineer anything with agents, from app behavior down to native binaries

Kevin Kelly's 2016 AI Predictions, 10 Years Later

How Anthropic’s Dynamic Workflows Coordinate Claude Agents

GitHub Copilot app for Beginners: How to teach Copilot about your codebase

Once again anthropic doing free marketing for open weight models

It's been 2 months

Qwen3.8-27B vs Frontier vs Others

Accessing OpenAI GPT models on Amazon Bedrock with global cross-Region inference in Thailand, Malaysia, Singapore, Indonesia, and Taiwan

Satya Nadella says we should assume all AI models are ‘compromised’

In 2030 owning a gpu cluster will be seen the same as owning a gun for home defense

Anthropic's Claude AI submitted false murder tip to Philadelphia police

arstechnica.com

LLC: lightweight Open WebUI alt now with Agents+MCP+RAG+Image Editing

Honeyjar offers AI tools to improve PR and comms at every company

Meta releases its first frontier model, Muse Wonder

WindBorne wins a hefty US Air Force contract for its balloon data

Meta has some Wonder Full new AI models

vllm-ascend updates (GLM5.3-flash) on dual 310p Ascend cards

Import AI 476: Creativity and RSI; worse-than-China open weight models; AI goes to space

importai.substack.com

Shipping Media Over QUIC at Meta Scale for Instagram Live Ingest

engineering.fb.com

Turning Opportunistic AI Inference Capacity Into Better Ads Recommendations

engineering.fb.com

Simulating student learning for better AI tutoring

UPDATE: Qwen 3.8 27B 140 tok/s on single RTX 3090 Megakernel: KL divergence 0.0009 vs llama.cpp

Filter, reshape, and redact logs before export with Transformers

blog.cloudflare.com

Deploy decision models with llama.cpp on Amazon SageMaker AI

Run multi-node inference workloads using Ray LLM Deep Learning Containers

Boole-1.7B: A 1-bit Qwen3-1.7B model that outperforms Bonsai-1.7B

graphsignal-top: a `top` for inference GPUs in the open-source Graphsignal profiler, kernels ranked by time, live, with readable names

Microsoft president says UK is best placed to drive AI regulation

This AI startup says it can generate permit-ready blueprints in minutes

← All editions