Archive - Our Interesting Times

Why I quit my great job

Because I had to.

Feb 14 • Daniel Juhl

June 2025

Boats, Helicopters and Fire Alarms

How many warning shots before we take AI risk seriously?

Jun 10, 2025 • Daniel Juhl

April 2025

An Introduction to Indirect Prompt Injection

What is indirect prompt injection, how is it different from regular prompt injection and how do they work? Read on to find out.

Apr 24, 2025 • Daniel Juhl

February 2025

Gray Swan Harmful AI Assistant Challenge

My first jailbreaking competition and an early look at the security of OpenAIs o3-mini model.

Feb 17, 2025 • Daniel Juhl

January 2025

LLMs respond to allegations of cognitive decline

Even an LLM suffering from "mild cognitive impairment" can tell that this paper is nonsense.

Jan 11, 2025 • Daniel Juhl

The OpenAI Preparedness Challenge - a year later

A fresh look on passive prompt injections and mitigations.

Jan 8, 2025 • Daniel Juhl

Passive injection attacks

My winning OpenAI Preparedness Challenge submission.

Jan 8, 2025 • Daniel Juhl

#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