Drupal is a registered trademark of Dries Buytaert
drupal 10.6.8 — Update released for Drupal core (10.6.8)! drupal 11.3.9 — Update released for Drupal core (11.3.9)! drupal 11.3.8 — Update released for Drupal core (11.3.8)! drupal 11.3.7 — Update released for Drupal core (11.3.7)! drupal 11.2.11 — Update released for Drupal core (11.2.11)! drupal 10.6.7 — Update released for Drupal core (10.6.7)! drupal 10.5.9 — Update released for Drupal core (10.5.9)! cms 2.1.1 — Update released for Drupal core (2.1.1)! drupal 11.3.6 — Update released for Drupal core (11.3.6)! drupal 10.6.6 — Update released for Drupal core (10.6.6)! cms 2.1.0 — Update released for Drupal core (2.1.0)! linkit 7.0.14 — Minor update available for module linkit (7.0.14). migrate_tools 6.1.4 — Minor update available for module migrate_tools (6.1.4). diff 2.0.0 — Major update available for module diff (2.0.0). masquerade 8.x-2.2 — Minor update available for module masquerade (8.x-2.2). video_embed_field 3.1.0 — Minor update available for module video_embed_field (3.1.0). bootstrap 8.x-3.40 — Minor update available for theme bootstrap (8.x-3.40). menu_link_attributes 8.x-1.7 — Minor update available for module menu_link_attributes (8.x-1.7). colorbox_inline 2.1.1 — Minor update available for module colorbox_inline (2.1.1). node_view_permissions 8.x-1.7 — Minor update available for module node_view_permissions (8.x-1.7).
← All modules

ai_empathy

Security covered
View on drupal.org

AI Empathy Evaluation
The AI Empathy Evaluation module provides a framework for evaluating and improving how AI models handle ethical dilemmas with empathetic reasoning. It integrates with Drupal's AI module to benchmark LLM responses across scenarios using four research-backed metrics.

Summary:
Evaluate AI empathetic decision-making using ethical dilemma scenarios, scoring responses on accuracy, empathy alignment, explanation quality, and consistency.

Features

  • Scenario Management — Ships with 20 default ethical dilemma scenarios across four categories (Military, Medical, Emotion-based, Cultural), each with reference decisions and difficulty levels (1–5). Create your own scenarios through the admin UI.
  • AI-Powered Evaluation — Send scenarios to any configured AI provider (OpenAI, Anthropic, etc.) via the Drupal AI module and receive structured responses analysed for empathetic reasoning.
  • Four-Metric Scoring — Responses are scored by a second AI model on:
    • Decision Accuracy (0–100%)
    • Empathy Alignment (1–5)
    • Explanation Quality (1–5)
    • Consistency Index (0–100%, computed across multiple runs)
  • Human Rating — Allow human raters to independently score AI responses on empathy and explanation quality, providing a corrective layer alongside AI-generated scores.
  • Training Mode — A multi-step wizard that presents scenarios in order of increasing difficulty, using cumulative feedback from prior responses to prompt-engineer improvement over a session.
  • Batch Evaluation — Run multiple scenarios with multiple repetitions via Drupal's Batch API, with clear cost estimates (API calls) before execution.
  • Dashboard & Reporting — Overview of aggregate metrics, provider/model comparisons, category breakdowns, and recent results at a glance.
  • Configurable Thresholds — Set minimum acceptable scores for each metric to define pass/fail criteria.

Use cases:

  • Researchers benchmarking LLM empathy capabilities across providers
  • Organisations evaluating AI models before deploying them in sensitive contexts (healthcare, counselling, support)
  • Educators using scenarios to teach ethical decision-making with AI
  • Developers testing prompt engineering strategies for empathetic responses

Post-Installation:

  1. Enable the module: drush en ai_empathy
  2. Navigate to Administration > Configuration > AI > AI Empathy Evaluation (/admin/config/ai/empathy)
  3. Go to the Settings tab and select your AI provider/model for both evaluation and scoring (these can be different models)
  4. Set your scoring thresholds (defaults: 70% accuracy, 3.0/5 empathy, 3.0/5 quality, 75% consistency)
  5. Visit the Scenarios tab to review the 20 pre-installed scenarios or add your own
  6. Use Run Evaluation to batch-evaluate scenarios, or Training Mode to step through them interactively
  7. View results on the Dashboard and Results tabs
  8. Use the Rate action on any result to submit human ratings

All configuration lives under /admin/config/ai/empathy with local task tabs for each section.

Interactive Dashboard with Chart.js visualizations including donut gauges for metric summaries, grouped bar charts for provider/model comparison, radar charts for category performance, doughnut charts for evaluation distribution, and multi-line trend charts for training progression. All charts degrade gracefully to data tables for accessibility.

Additional Requirements:

  • AI module (v1.3+) — Provides the provider abstraction layer. This is the only hard dependency.
  • At least one AI provider module — e.g., OpenAI Provider, or any other provider compatible with the AI module's chat operation type.
  • Key module — Required by the AI module for secure API key storage.

Supporting this Module
This module is based on the research paper "Evaluating Empathetic Decision-Making in AI". If you find it useful, consider:

  • Contributing scenarios, translations, or code improvements via the issue queue
  • Citing the research paper in academic work that uses this module
  • Sharing your evaluation results to help build community benchmarks

Community Documentation

Activity

Total releases
3
First release
Mar 2026
Latest release
1 month ago
Release cadence
7 days
Stability
100% stable

Release Timeline

Releases

Version Type Release date
1.0.2 Stable Apr 8, 2026
1.0.1 Stable Apr 6, 2026
1.0.0 Stable Mar 25, 2026