The LLM warnings Google fired Timnit Gebru over have all come true

I built a vulnerable app and spent $1,500 seeing if LLMs could hack it

Companies Are Using Reddit to Manipulate ChatGPT and Google AI Search. Peptide companies have been doing AI-engine optimization by spamming the biohackers subreddit to manipulate ChatGPT and Google.

Artificial intelligence is not conscious – Ted Chiang

AI outperforms law professors in Stanford Law study

LLMs are not the black box you were promised

Gmail thinks I'm stupid, so I left

CS336: Language Modeling from Scratch

When AI Crosses the Line: The Matplotlib Incident

LLMs Are Closer to Religion Than They Appear

It's Not Just X. It's Y

Backpressure is all you need

EY Canada published a cybersecurity report and most citations were hallucinated

Expertise in the age of AI

Real-time LLM Inference on Standard GPUs: 3k tokens/s per request

The mysterious Hy3 LLM is topping OpenRouter Model Rankings by a large margin

Various LLM Smells

About LLMs at Zig Days

Claude Opus 4.8

Disagreement among frontier LLMs on real-world fact-checks

I think Anthropic and OpenAI have found product-market fit

Ripgrep AI Policy

I'm Tired of Talking to AI

Where does next-token prediction leave us?

From Rust to Ruby

Stack Overflow’s forum is dead but the company’s still kicking

A sleep-like consolidation mechanism for LLMs

Outsourcing plus local AI will soon become more economical vs. frontier labs

Using AI to write better code more slowly

Norway's 2 petabytes of Huawei flash storage and LLM training

Arena AI Model ELO History

The other half of AI safety

Rars: a Rust RAR implementation, mostly written by LLMs

“I applied to be pope”: Losing grip on reality while using ChatGPT

How Fast Does Claude, Acting as a User Space IP Stack, Respond to Pings?

Chrome's AI features may be hogging 4GB of your computer storage

LLMorphism: When humans come to see themselves as language models

LLMs Corrupt Your Documents When You Delegate

A recent experience with ChatGPT 5.5 Pro

Court to DOGE: Asking ChatGPT 'Is This DEI?' Is Not Proper Legal Process

Can LLMs model real-world systems in TLA+?

AMD's local, open-source AI can now easily interact with your Gmail

GPT-5.5 Price Increase: What It Costs

Natural Language Autoencoders: Turning Claude's Thoughts into Text

DeepSeek 4 Flash local inference engine for Metal

Making LLM Training Faster with Unsloth and NVIDIA

Appearing Productive in the Workplace

GPT‑5.5 Instant

Accelerating Gemma 4: faster inference with multi-token prediction drafters

Three Inverse Laws of AI