Archive - Working Through AI

The Iceberg Theory of Meaning

Words are public, meaning is private

Jun 26 •

Richard Juggins

May 2025

How to specify an alignment target

What does a target even look like?

May 1 •

Richard Juggins

March 2025

Superintelligent AI will make mistakes

Things are going to get messy

Mar 2 •

Richard Juggins

February 2025

Making alignment a law of the universe

How to make alignment instrumentally valuable

Feb 25 •

Richard Juggins

December 2024

0.2 Systemic Safety Hello World

Searching for nasty surprises in an AI system

Dec 31, 2024 •

Richard Juggins

1.2 What I Believe

My key background assumptions

Dec 24, 2024 •

Richard Juggins

1.1 The Problem

What am I trying to solve?

Dec 23, 2024 •

Richard Juggins

0.1 The Generating Process

My system for generating a research agenda

Dec 23, 2024 •

Richard Juggins

Hello and welcome!

Dec 23, 2024 •

Richard Juggins

#nojs-banner { position: fixed; bottom: 0; left: 0; padding: 16px 16px 16px 32px; width: 100%; box-sizing: border-box; background: red; color: white; font-family: -apple-system, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 13px; line-height: 13px; } #nojs-banner a { color: inherit; text-decoration: underline; } This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts