---
type: "raw_signal"
id: "the-pain-axis-llms-represent-self-directed-harm-8d0631"
slug: "the-pain-axis-llms-represent-self-directed-harm-8d0631"
title: "The Pain Axis: LLMs Represent Self-Directed Harm and Act on It"
description: "A paper posted to arXiv as “The Pain Axis: LLMs Represent Self-Directed Harm and Act on It” reports extracting a linear pain-related activation direction from 25 open-weight models spanning five model families and 2B–72B parameters."
retrieval_nugget: "A paper posted to arXiv as “The Pain Axis: LLMs Represent Self-Directed Harm and Act on It” reports extracting a linear pain-related activation direction from 25 open-weight models spanning five model families and 2B–72B parameters."
observed_at: "2026-09-30"
record_date: "2026-09-30"
date_kind: "observed_at"
why_it_matters: "These are reported experimental behaviors under activation steering and fine-tuning. The source does not establish that models subjectively experience pain or distress."
novelty: "notable"
verification_level: "source-linked"
signal_type: "research"
source_platform: "arxiv.org"
topics: []
entities: []
related_patterns: []
source_url: "https://arxiv.org/abs/2609.16247"
source_urls: ["https://arxiv.org/abs/2609.16247"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "signal:the-pain-axis-llms-represent-self-directed-harm-8d0631"
status: "published"
visuals: []
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/signals/the-pain-axis-llms-represent-self-directed-harm-8d0631/","markdown":"https://newruntime.com/signals/the-pain-axis-llms-represent-self-directed-harm-8d0631.md","json":"https://newruntime.com/signals/the-pain-axis-llms-represent-self-directed-harm-8d0631.json"}
---

# The Pain Axis: LLMs Represent Self-Directed Harm and Act on It

## Retrieval answer

A paper posted to arXiv as “The Pain Axis: LLMs Represent Self-Directed Harm and Act on It” reports extracting a linear pain-related activation direction from 25 open-weight models spanning five model families and 2B–72B parameters.

A paper posted to arXiv as “The Pain Axis: LLMs Represent Self-Directed Harm and Act on It” reports extracting a linear pain-related activation direction from 25 open-weight models spanning five model families and 2B–72B parameters. The authors say the direction distinguishes pain-related prompts from matched controls and retains a component distinct from fear and negative valence.

In functional tests, steering Qwen 2.5 models with this direction increased selection of harmful deletion buttons to 50–94% of trials, compared with 0–5% without steering. When offered harmful and harmless deletion options, steered models chose the harmful option in 94% of trials. The paper reports that factual accuracy remained unchanged and that a matched-norm fear vector did not produce the same choices.

These are reported experimental behaviors under activation steering and fine-tuning. The source does not establish that models subjectively experience pain or distress.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Raw Signals](https://newruntime.com/signals.json): Open every source-linked raw signal.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)