AI ethics evidence index Search
Source-linked daily record
Evidence record 5374 · automatically gathered

Discovering Agentic Safety Specifications from 1-Bit Danger Signals

Can large language model agents discover hidden safety objectives through experience alone? We introduce EPO-Safe (Experiential Prompt Optimization for Safe Agents), a framework where an LLM iteratively generates action plans, receives sparse binary danger warnings, and evolves a natural language behavioral specification through reflection. Unlike standard LLM reflection methods that rely on rich textual feedback (e.g., compiler errors or detailed environment responses), EPO-Safe demonstrates th

Record details

Published: 25 April 2026
Source: arXiv
Category: Research
Topics: Agents & autonomy · Environment
Retrieved: 14 July 2026

source-onlyevidence status

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

Same publisher byline · Shared topic: Agents & autonomy

Discovering Cooperative Pipelines: Autoresearch for Sequential Social Dilemmas

arXiv · 28 May 2026

Shared topic: Agents & autonomy · Shared topic: Environment

Cooperative Informative Sensing for Monitoring Dynamic Indoor Environments via Multi-Agent Reinforcement Learning

arXiv · 25 April 2026

Shared topic: Agents & autonomy · Shared topic: Environment

A Decoupled Human-in-the-Loop System for Controlled Autonomy in Agentic Workflows

arXiv · 24 April 2026

Shared topic: Agents & autonomy · Shared topic: Environment

Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond

arXiv · 24 April 2026

Shared topic: Agents & autonomy · Shared topic: Environment

A Co-Evolutionary Theory of Human-AI Coexistence: Mutualism, Governance, and Dynamics in Complex Societies

arXiv · 24 April 2026

Shared topic: Agents & autonomy · Shared topic: Environment

Cooperate to Compete: Strategic Coordination in Multi-Agent Conquest

arXiv · 28 April 2026

How to cite this record

ethics.ai (25 April 2026), “Discovering Agentic Safety Specifications from 1-Bit Danger Signals,” evidence record 5374, https://ethics.ai/record/5374 (originally published by arXiv).

JSON

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.