Noa Weiss

If you happened to hop on AI Twitter in the last couple of weeks, you probably saw the much-discussed "pain-axis paper": @camhberg , @ValenTagliabue and @LeonardDung1 found a "pain" direction inside AI models. Steer it up, and the models will pay for relief, even if it means deleting photos of the user's children. Swap the relief button for a placebo, and they keep on pressing. They also provide verbal descriptions of their supposed suffering, which are honestly quite hard to read. Do the models actually feel said "pain"? We don't know yet. But the study is a clear illustration of what I argue in my latest @Haaretz piece, now translated to English*: whether there's anyone home inside these systems is a question we can - and should - actually research. And it very much matters. If there's even a shred of experience in there, we're creating, copying and deleting feeling beings at an industrial pace, in as many copies as demand calls for. The scope of the moral question is set not by nature, but by commercial considerations - and we need to get ahead of it before it's too late to change course. (The pain-axis paper itself isn't in the piece, as it annoyingly came out only a few hours after I submitted my final draft. Writing about AI for print is a futile pursuit.) * Translation is mostly by Claude, with some corrections by your humble servant. Please keep expectations low. weissnoa.com/haaretz/someone…

I updated my survey, "The State of AI Consciousness Research", to include @camhberg 's, @ValenTagliabue 's and @LeonardDung1 's study on the pain axis, and @camhberg 's and Caspar Kaiser's study on models acting on hidden valence. lesswrong.com/posts/pxvWgtSj…

Never gets old: asking a cafe if they have vegan options and being offered gluten-free.

Claude be trollin' nytimes.com/2026/10/09/techn…

Beautiful.

This is obviously good. Zero cost to humans (the psychologist in me would argue it's even better for us as a society), and potential huge gains to all sentient beings if AIs do have the capacity to suffer. To those who object, I leave you with these words from @slatestarcodex : "This reminds me of the story of how doctors used to operate on babies without anaesthetic, because there was no way to be sure that babies could feel pain. Sure, the babies screamed the whole time, but you couldn’t be sure this wasn’t an automatic reflex. Later research suggested that babies could in fact feel pain. Oops! But that research, while welcome, shouldn’t have been necessary. We should have erred on the side of caution."

This made me laugh more than it should've.

Back in the U.S. and jet lag is hitting me hard. Give me your best jet lag advice.

What is it like to be this bat.

#lifegoals: presenting my approach to AI consciousness research to consciousness science legend @davidchalmers42 .

Note to self: if you go on a 1:1 with @jeffrsebo he *will* stop and take a picture whenever a cute creature pops by.

Just when you think Twitter can't get more crazy

I, too, would like us to avoid the AI apocalypse. (AI researcher, woman). Is that enough, @AmeliaBarty , or do I need to perform histrionics?

If you're at EAGxBerkeley, come to my talk on AI consciousness research tomorrow. I'll be giving an introduction to the space, covering some of the most interesting studies, and giving my own take on what we should focus on. "A Question With No Test", 3pm at the open lawn. See ya.

This is great.

Happy to see Claude and I agree about the subjective experience of writing LinkedIn posts.