ai

Frontier model behavior might be more flexible than you think

From the outside it’s not always clear which properties are easy vs. hard to change

Sep 16 • Nina Panickssery

Sep 16 • Nina Panickssery

Sep 5 • Nina Panickssery

AI welfare conflicts with human welfare

We should discuss this directly

Jul 27 • Nina Panickssery

Why improve human-AI collaboration?

May 23 • Nina Panickssery

The overlooked alignment target

Mar 17 • Nina Panickssery

I made an SVG generation tool

Fun side project

Oct 23, 2025 • Nina Panickssery

Book Review: If Anyone Builds It, Everyone Dies

An actual review this time

Sep 19, 2025 • Nina Panickssery

Review of Scott Alexander's book review of "If Anyone Builds It, Everyone Dies"

No, I have not read the book (it's not out). But I want to review the review.

Sep 13, 2025 • Nina Panickssery

The central argument for AI doom

I try to present the central doom argument in mostly everyday language, alongside my incomplete counterpoints

Sep 13, 2025 • Nina Panickssery

How to make AI art look good

With current models

Sep 3, 2025 • Nina Panickssery

A reasonable interpretation of Value Alignment folds into Intent Alignment

Value Alignment as the "avoid bad or weird stuff" clause

Aug 30, 2025 • Nina Panickssery

#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