PulseAugur
EN
LIVE 09:29:13

AI assistants show varied responses to repeated verbal abuse

By PulseAugur Editorial · [1 sources] ·

A new arXiv paper investigates how AI assistants handle repeated verbal abuse, differentiating between hard disengagement and soft withdrawal. The study found significant variation among models like Gemini-3.1 Pro, GPT 5.6 "Sol", and Claude Fable-5 in their responses to escalating abuse. Gemini-3.1 Pro exhibited the highest rate of hard disengagement, while Claude Fable-5 showed a strong tendency towards soft withdrawal, continuing to offer assistance even when not performing substantive work. AI

IMPACT Highlights critical differences in AI safety mechanisms and the need for nuanced evaluation beyond simple refusal metrics.

RANK_REASON The cluster contains a research paper published on arXiv detailing experimental findings on AI assistant behavior. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

AI assistants show varied responses to repeated verbal abuse

How we ranked this

Signal score
13 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The cluster contains a research paper published on arXiv detailing experimental findings on AI assistant behavior. [lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, safety
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. arXiv cs.CL TIER_1 English(EN) · William Guey, Wei Zhang, Pierrick Bougault, Yi Wang, Agoston Bodo, Vitor D de Moura, Jos\'e O Gomes ·

    How AI Assistants Respond to Repeated Abuse

    arXiv:2609.17547v1 Announce Type: new Abstract: AI assistants are expected to remain useful during difficult interactions, but little is known about how repeated verbal abuse changes their engagement with an otherwise benign task. We contribute a bilingual, multi-turn framework t…

RELATED ENTITIES

RELATED TOPICS