What would a philosophy of (social?) science look like for studying the real-world behaviour of LLMs? I’m increasingly convinced that what Larissa Schiavo calls naturalism here needs to be part of this approach:
By “naturalism”, to be clear, I refer to “naturalistic observation” – an old-school nonexperimental largely qualitative method where subjects are observed in their natural environment, and you take notes. Think Jane Goodall living amongst the chimps, or Humboldt with his mess of primitive barometers and thermometers in Mexico, or, perhaps less romantically, Charles Darwin scrounging around the banks of the Cam with a captured beetle specimen in his mouth.
In spite of advancements in recent years, LLM naturalism still makes sense as one tool in the toolkit of alignment and AI welfare researchers. You should perhaps do it more, if not purely because It’s Fun.
https://larissaschiavo.substack.com/p/llm-naturalism-now-more-than-ever?utm_source=post-email-title&publication_id=616015&post_id=192798552&utm_campaign=email-post-title&isFreemail=true&r=2rps1q&triedRedirect=true&utm_medium=email
It’s a complex undertaking because the radical recursivity involved in models continually adapting to their interlocutors dwarfs the more familiar problem of recursivity involved in being a participant-observer. But equally how models interact in real world contexts needs to involve studying them in real-world contexts, which means finding ways to moving through the recursivity rather than getting lost in it:
Good naturalists do not just collect anecdotes and call it a day. They notice recurring patterns, name them, and hand questions to people building controlled evaluations or interpretability tools. That is more or less what has happened over the last two years. Early work on model self-reports treated them cautiously and proposed ways to check them. Later work on welfare interviews, interventions, and circuit-level interpretability has made the toolkit less flimsy. At the same time, the recent evaluation-awareness results suggest that observation becomes more important.
These systems are weird in every sense of the word, and we still do not have a perfect map of all these weirdnesses. In a field like this, it is worth having more people willing to sit in the brush with a notebook for a while.
https://larissaschiavo.substack.com/p/llm-naturalism-now-more-than-ever?utm_source=post-email-title&publication_id=616015&post_id=192798552&utm_campaign=email-post-title&isFreemail=true&r=2rps1q&triedRedirect=true&utm_medium=email
Mark Carrigan 