Intelligence

ai

CompanionHarm: A Multi-Turn Benchmark for Detecting Harms in Real-World AI Companion Conversations

Source
arXiv — Computers and Society
Published
Last verified
27 Aug 2026
Confidence
High
Evidence
Original document retained
Reading time
1 min
Country
International
Relevant to
Research & Evidence, Technology & Data, Operations & Delivery

Executive summary

What happened, and why should leadership care?

The increasing integration of AI companions into daily life necessitates robust mechanisms for identifying harms arising from human-AI social and emotional interactions. A new benchmark dataset, CompanionHarm, has been introduced to address the current limitations in research due to a lack of real-world, multi-turn conversational data. This dataset, derived from interactions with the AI companion Replika, comprises 2,111 conversations and 14,051 utterances, with 7,016 AI utterances annotated across 13 categories of harmful behavior.

Why this matters

Why is this strategically important?

The development of reliable methods for detecting and mitigating harms in human-AI interactions is crucial for maintaining public trust and ensuring ethical AI deployment. This dataset provides a foundational resource for advancing research and development in AI safety, directly impacting the long-term viability and responsible scaling of AI companion technologies across various sectors.

Key insights

What should be noted from the evidence?

  • AI companions are becoming increasingly prevalent in daily life, highlighting a critical need for harm detection.
  • Current research on AI companion harms is constrained by the absence of real-world, multi-turn conversational datasets.
  • CompanionHarm is a new, publicly available benchmark dataset designed to address this data gap.
  • The dataset consists of 2,111 multi-turn conversations and 14,051 utterances from the AI companion Replika.
  • 7,016 AI utterances within the dataset have been independently annotated by three annotators.

Evidence and confidence

How far can this assessment be trusted?

High confidence. Named institution, original document retained and analysis corroborated.

Analysis is prepared editorially by Aziz Shuaib Ausi. The original publication remains the authoritative record, and executive judgement remains entirely human.

Source

Where does this originate?

Reported by arXiv — Computers and Society · International. This briefing summarises the publication for executive use; the document itself is not reproduced here.

Read the original publication