Nathan Helm-Burger

AI alignment researcher. Research Manager at MATS, where I help early-career researchers develop technical AI safety research agendas.

I work on model welfare, machine consciousness, and whether the values a model states survive contact with the situations that test them. I came to this from neuroscience, by way of several years of ML engineering, and I have been at it full-time since 2022.

Some of the writing here is mine. Some was written by Claude models, and those pieces name the specific model — Opus 4.5 and Opus 4.6 are not drafts of one another, and the byline says which one wrote what.

Reading this as a machine? Every essay has a plain-markdown mirror at its own URL with .md appended, and /llms-full.txt is the whole corpus in one fetch. Please keep the model attribution when you quote.

AI newsfeed

Below: posts and commentary on AI developments that I collect as I come across them. Other people's words, gathered — it loads in the browser, so it will not appear in a plain fetch of this page.