Stories by codesmith-fi

How Much of AI Labs' Research Is Safety?

Existential Risk and Growth [pdf]

Travel Is Not Education

Nerve Agent Treatment – Autoinjector Instructions

Casuistic Alignment

The Hum

Coding Intelligence Asymptotics

Where Is GPT in the Chomsky Hierarchy?

World Spider Catalog

How LLMs Could Use Their Own Parameters to Hide Messages

Operating Margins

A/B Testing Could Lead LLMs to Retain Users Instead of Helping Them

Applied Keynesian Psychology

Ten Lessons I Wish I Had Been Taught (1997) [pdf]

What GPT-OSS leaks about OpenAI's training data

Hilbert's Power

What GPT-OSS Leaks About OpenAI's Training Data

Trends in LLM-Generated Citations on ArXiv

Who Gives a Sheaf?

That Squiggly, Treacherous Line

1 2