·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Romanized Arabic Across Dialects: Views, Usage Patterns, and Linguistic Variation4h◆Do AI-for-Science Agents Preserve Model Safety?4h◆Can Test-Time Memory Evolution Outperform Test-Time Training for Scientific Discovery?4h◆HyScale: Scaling Long-Context Training for Learned Sparse Attention4h◆Understanding Communication Failures in Multi-Agent LLM Systems4h◆DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving4h◆Behaviour analysis for large language model-driven agents4h◆Who Verifies the Tools? How Warped Formatters Mislead Agents and Their Verifiers4h◆MPQ: A Message-Passing View of Post-Training Quantization4h◆$\Psi$-0.6: Adaptive Streams of Thought in a Unified Vision-Language World Model4h◆Not All Information Is Worth Asking For: Decision-Informed Information Seeking with LLMs4h◆Evidence-Grounded SLM Behavioral Prediction Architecture for Future Assistance Use4h◆SWE-Shift: Evaluating Coding Agents in Evolving Workspaces4h◆Solvent: Long-Horizon Reasoning for Multi-Step Retrosynthesis4h◆When to Stop Penalizing Actions? Martingale GRPO for VLA Reinforcement Learning4h◆POSEIDON: Scaling Training Environments for Scientific Agents from Code Repositories4h◆ALLMTS: Automated LLM-based Medical Triage System4h◆When Agents Fail: Model or Harness--Who's to Blame?4h◆LGMT: A Tool for Testing the Logical Reasoning Reliability of LLMs4h◆WiMoT: Replay-Free Continual Wi-Fi Sensing via Moment Transport4h◆Romanized Arabic Across Dialects: Views, Usage Patterns, and Linguistic Variation4h◆Do AI-for-Science Agents Preserve Model Safety?4h◆Can Test-Time Memory Evolution Outperform Test-Time Training for Scientific Discovery?4h◆HyScale: Scaling Long-Context Training for Learned Sparse Attention4h◆Understanding Communication Failures in Multi-Agent LLM Systems4h◆DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving4h◆Behaviour analysis for large language model-driven agents4h◆Who Verifies the Tools? How Warped Formatters Mislead Agents and Their Verifiers4h◆MPQ: A Message-Passing View of Post-Training Quantization4h◆$\Psi$-0.6: Adaptive Streams of Thought in a Unified Vision-Language World Model4h◆Not All Information Is Worth Asking For: Decision-Informed Information Seeking with LLMs4h◆Evidence-Grounded SLM Behavioral Prediction Architecture for Future Assistance Use4h◆SWE-Shift: Evaluating Coding Agents in Evolving Workspaces4h◆Solvent: Long-Horizon Reasoning for Multi-Step Retrosynthesis4h◆When to Stop Penalizing Actions? Martingale GRPO for VLA Reinforcement Learning4h◆POSEIDON: Scaling Training Environments for Scientific Agents from Code Repositories4h◆ALLMTS: Automated LLM-based Medical Triage System4h◆When Agents Fail: Model or Harness--Who's to Blame?4h◆LGMT: A Tool for Testing the Logical Reasoning Reliability of LLMs4h◆WiMoT: Replay-Free Continual Wi-Fi Sensing via Moment Transport4h◆
News/Are LLMs Bad at Moral Reasoning?
arxiv
PublishedJune 11, 2026 at 4:00 AM
▲bullish

Are LLMs Bad at Moral Reasoning?

Source
arxiv.orgfull article ↗
Read on arxiv→
Publisher summary· verbatim

arXiv:2606.11635v1 Announce Type: cross Abstract: For highly capable AI systems to operate safely in dynamic, open-ended environments, they must be able to identify, understand, and respond to moral reasons for action, and constrain their behaviour accordingly. A growing body of research aims to eva

Stay posted· Newsletter

Three short emails a week: Monday, Wednesday and Friday. Top movers, new models and the most-referenced story.

// no spam · unsubscribe one-click · free forever

Your email is used only to send The Bubble Brief. Privacy policy

Discussion
Mentioned models
01
  • 01
    LLMs
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
03
#moral-reasoning#ai-safety#evaluation

No replies yet. Be first.

Mentioned models
01
  • 01
    LLMs
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
03
#moral-reasoning#ai-safety#evaluation

Related coverage

More from ARXIV
arxivRomanized Arabic Across Dialects: Views, Usage Patterns, and Linguistic Variation4harxivDo AI-for-Science Agents Preserve Model Safety?4harxivCan Test-Time Memory Evolution Outperform Test-Time Training for Scientific Discovery?4harxivHyScale: Scaling Long-Context Training for Learned Sparse Attention4h
The Bubble Brief
MON · WED · FRI

Read moral-reasoning insights every Monday, Wednesday and Friday — top movers, new models, story of the week.

// no spam · unsubscribe one-click · free forever

Your email is used only to send The Bubble Brief. Privacy policy

Originally published on arxiv ↗
PrivacyTermsRefundsLegal notice|AboutMethodologyContact

© 2026 Databubble

Built by Marouane Gazouzi · Readyo, AI interview coach

HomeModelsNews