arbiter
JournalsAccount

Helpful, harmless, honest? Sociotechnical limits of AI alignment and safety through Reinforcement Learning from Human Feedback

Adam Dahlgren Lindström et al.

Ethics and Information Technology2025https://doi.org/10.1007/s10676-025-09837-2article
AJG 1ABDC B
Weight
0.61

Abstract

This paper critically evaluates the attempts to align Artificial Intelligence (AI) systems, especially Large Language Models (LLMs), with human values and intentions through Reinforcement Learning from Feedback methods, involving either human feedback (RLHF) or AI feedback (RLAIF). Specifically, we show the shortcomings of the broadly pursued alignment goals of honesty, harmlessness, and helpfulness. Through a multidisciplinary sociotechnical critique, we examine both the theoretical underpinnings and practical implementations of RLHF techniques, revealing significant limitations in their approach to capturing the complexities of human ethics, and contributing to AI safety. We highlight tensions inherent in the goals of RLHF, as captured in the HHH principle (helpful, harmless and honest). In addition, we discuss ethically-relevant issues that tend to be neglected in discussions about alignment and RLHF, among which the trade-offs between user-friendliness and deception, flexibility and interpretability, and system safety. We offer an alternative vision for AI safety and ethics which positions RLHF approaches within a broader context of comprehensive design across institutions, processes and technological systems, and suggest the establishment of AI safety as a sociotechnical discipline that is open to the normative and political dimensions of artificial intelligence.

15 citations

Open via your library →

Cite this paper

https://doi.org/https://doi.org/10.1007/s10676-025-09837-2

Or copy a formatted citation

@article{adam2025,
  title        = {{Helpful, harmless, honest? Sociotechnical limits of AI alignment and safety through Reinforcement Learning from Human Feedback}},
  author       = {Adam Dahlgren Lindström et al.},
  journal      = {Ethics and Information Technology},
  year         = {2025},
  doi          = {https://doi.org/https://doi.org/10.1007/s10676-025-09837-2},
}

Paste directly into BibTeX, Zotero, or your reference manager.

Flag this paper

Helpful, harmless, honest? Sociotechnical limits of AI alignment and safety through Reinforcement Learning from Human Feedback

Flags are reviewed by the Arbiter methodology team within 5 business days.

Evidence weight

0.61

Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40

F · citation impact0.63 × 0.4 = 0.25
M · momentum0.88 × 0.15 = 0.13
V · venue signal0.50 × 0.05 = 0.03
R · text relevance †0.50 × 0.4 = 0.20

† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.