Why don't people use formal methods? (2019)

Some thoughts about Anthropic's new cryptanalysis results

Document-borne AI worms can self-propagate through Copilot for Word

Anthropic publishes a practical key-recovery attack on HAWK-256

Codex Security

Pacing the frontier

Discovering Cryptographic Weaknesses with Claude

Fast Remediation Is the New Trust Model (JFrog and OpenAI Zero-Day Findings)

Google's Beyond Zero: Enterprise Security for the AI Era

Don't ask an LLM for a confidence score

Our position on open-weights models

This July I Was Fired from Simple AI (A Deeply YC Company)

The Dark Night of Mathematics

AIs don't do what you want. This is bad

Be skeptical of OpenAI's rogue hacker agent story

A taxonomy of omnicidal futures involving artificial intelligence (2025)

What happened to TheNumbers.com

The arguments against open source AI are bad

I Think You Might Be Fooling Yourself with AI

OpenAI and Anthropic unite against open-weight AI risks to their bottom line

OpenAI’s accidental attack against Hugging Face is science fiction that happened

OpenAI says its AI went rogue and launched 'unprecedented' cyber-attack

Codeberg bans vibe coded projects

OpenAI and Hugging Face address security incident during model evaluation

Annoying and alarming things about OpenCode

Our Approach to Bioresilience: Isomorphic Labs and Google DeepMind

Homomorphically encrypted CIFAR-10 inference in 200ms

Decoy Font

Towards a Harness That Can Do Anything

I tricked Claude into leaking your deepest, darkest secrets

Microsoft Patches a Record 570 Security Flaws

OpenAI mandates hardware-backed passkeys for Trusted Access Cyber members

Guardian Angels: LLM Personalization for Productivity and Security

Demis Hassabis has a plan to harness AI safely

The One-Step Trap (In AI Research)

AI 2040 and the cult of intelligence

A font that humans can read but AI cannot

An update on residential proxies and the scraper situation

Ben Bernanke Joins Anthropic Oversight Trust

Build your own vulnerability harness

AI 2040: Plan A

MIRA: Multiplayer Interactive World Models Trained on Rocket League

Chat Control passed first round in EU Parliament

What Emily Bender meant by "stochastic parrots"

New serious vulnerabilities spiked around release of Claude Mythos Preview

Alibaba to ban Claude Code in workplace over alleged backdoor risks, source says

EFF letter to FTC on X consent order (2 July 2026) [pdf]

The short leash AI coding method for beating Fable

No LLM Code in Dependencies

Anthropic has embedded hidden spyware-like code in Claude Code