Simon Willison's Weblog simonwillison.net

Cato CTRL™ Threat Research: PoC Attack Targeting Atlassian’s Model Context Protocol (MCP) Introduces New “Living off AI” Risk

simonwillison.net·45w

playbackrate

simonwillison.net·45w

How OpenElections Uses LLMs

simonwillison.net·45w

Clarified zucchini consommé

simonwillison.net·45w

Quoting Arvind Narayanan

simonwillison.net·45w

Quoting Workaccount2 on Hacker News

simonwillison.net·45w

coding-agents

simonwillison.net·45w

I counted all of the yurts in Mongolia using machine learning

simonwillison.net·45w

It's a trap

simonwillison.net·45w

Trying out the new Gemini 2.5 model family

simonwillison.net·45w

Quoting Donghee Na

simonwillison.net·45w

100% effective

simonwillison.net·45w

Cloudflare Project Galileo

simonwillison.net·45w

Quoting Paul Biggar

simonwillison.net·45w

The lethal trifecta for AI agents: private data, untrusted content, and external communication

simonwillison.net·46w·Lobsters, Hacker News, Hacker News, Hacker News

Quoting Joshua Barretto

simonwillison.net·46w

Seven replies to the viral Apple reasoning paper – and why they fall short

simonwillison.net·46w

An Introduction to Google’s Approach to AI Agent Security

simonwillison.net·46w

Anthropic: How we built our multi-agent research system

simonwillison.net·46w·Hacker News

llm-fragments-youtube

simonwillison.net·46w

Log in to enable infinite scrolling