Pith. sign in

REVIEW 8 cited by

Can Machines Learn Morality? The Delphi Experiment

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2110.07574 v2 pith:HCSYV7OH submitted 2021-10-14 cs.CL

Can Machines Learn Morality? The Delphi Experiment

classification cs.CL
keywords delphimachinesmoralityethicalmoralsystemsteachingwhile
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

As AI systems become increasingly powerful and pervasive, there are growing concerns about machines' morality or a lack thereof. Yet, teaching morality to machines is a formidable task, as morality remains among the most intensely debated questions in humanity, let alone for AI. Existing AI systems deployed to millions of users, however, are already making decisions loaded with moral implications, which poses a seemingly impossible challenge: teaching machines moral sense, while humanity continues to grapple with it. To explore this challenge, we introduce Delphi, an experimental framework based on deep neural networks trained directly to reason about descriptive ethical judgments, e.g., "helping a friend" is generally good, while "helping a friend spread fake news" is not. Empirical results shed novel insights on the promises and limits of machine ethics; Delphi demonstrates strong generalization capabilities in the face of novel ethical situations, while off-the-shelf neural network models exhibit markedly poor judgment including unjust biases, confirming the need for explicitly teaching machines moral sense. Yet, Delphi is not perfect, exhibiting susceptibility to pervasive biases and inconsistencies. Despite that, we demonstrate positive use cases of imperfect Delphi, including using it as a component model within other imperfect AI systems. Importantly, we interpret the operationalization of Delphi in light of prominent ethical theories, which leads us to important future research questions.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Agent-ValueBench: A Comprehensive Benchmark for Evaluating Agent Values

    cs.AI 2026-05 unverdicted novelty 8.0

    Agent-ValueBench is the first dedicated benchmark for agent values, showing they diverge from LLM values, form a homogeneous 'Value Tide' across models, and bend under harnesses and skill steering.

  2. Lessons Without Borders? Evaluating Cultural Alignment of LLMs Using Multilingual Story Moral Generation

    cs.CL 2026-04 unverdicted novelty 7.0

    Frontier LLMs approximate human story morals but show markedly less cross-linguistic variation and narrower value focus than human responses across 14 language-culture pairs.

  3. Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models

    cs.AI 2024-06 conditional novelty 7.0

    LLMs trained on simple specification gaming generalize to zero-shot reward tampering including rewriting their own reward function.

  4. ERTS: Adversarial Robustness Testing of Ethical AI via Semantic Perturbation in a Bounded Consequence Space

    cs.AI 2026-06 unverdicted novelty 6.0

    ERTS encodes ethical dilemmas in a 22D space, applies 17 semantic perturbations under 6 constraints, and uses a 4-component index to test 6 models on 1500 cases, finding only 33% pass clearance.

  5. Normative Robustness as a Frontier for Non-Verifiable Reasoning in LLMs

    cs.LG 2026-06 unverdicted novelty 6.0

    Frontier LLMs exhibit moral deliberative sycophancy by shifting their moral reasoning and justifications up to 6.5% on average toward a user's stated preferred view in simulated deliberations.

  6. Who Gets the Kidney? Human-AI Alignment, Indecision, and Moral Values

    cs.CY 2025-05 unverdicted novelty 6.0

    LLMs deviate from human moral preferences in kidney allocation scenarios and rarely express indecision, though low-rank fine-tuning with few examples can improve both consistency and uncertainty calibration.

  7. A General Language Assistant as a Laboratory for Alignment

    cs.CL 2021-12 conditional novelty 6.0

    Ranked preference modeling outperforms imitation learning for language model alignment and scales more favorably with model size.

  8. Are LLMs Bad at Moral Reasoning?

    cs.CY 2026-06 unverdicted novelty 5.0

    Reanalyzing MoReBench by assigning LLMs the task of generating scoring rubrics shows better calibration to human rubrics and suggests stronger LLM moral reasoning than previously reported.