Journal article · 2026 · Open access

AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework

Sofia Vei, Paolo Giudici, Pavlos Sermpezis, Athena I. Vakali, Adelaide Emma Bernardelli

Artificial Intelligence, 358, 104587 · Published 6 July 2026 · DOI 10.1016/j.artint.2026.104587

Open full textCite this paper

In brief

“AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework” is a 2026 journal article by Sofia Vei and 4 co-authors, published in Artificial Intelligence. OpenAlex counts 4 citations as of 18 November 2026, which puts it in the top 4% of works published in 2026 by citations. It is open access under a CC BY licence: the full text is free to read at Artificial Intelligence.

Citations
4
Most cited in
2026 (4)
Field-weighted impact
4.19
References
63
Authors
5

Abstract

We introduce AI Harmonics ( AIH ), a novel metric designed to quantify the concentration of harms across stakeholder groups affected by AI systems. Unlike traditional approaches that rely on arbitrary numerical assignments to ordinal severity levels, AIH provides a principled framework grounded in inequality theory, extending concepts from the Gini index to purely ordinal data. The metric evaluates how harm is distributed among stakeholders, capturing whether severe impacts are concentrated within specific groups or more evenly spread. Experiments on annotated incident data show that the proposed metric exhibits a strong monotonic relationship with the Criticality Index (CI), preserving harm category rankings while capturing additional variation in concentration patterns. The method demonstrates high robustness, with Spearman rank correlations above 0.97 under severity perturbations and stable prioritization even under up to 80% random data removal. Political and physical harms consistently exhibit the highest concentration, indicating the need for urgent mitigation. Political harms erode public trust, while physical harms pose serious, even life-threatening risks, underscoring the real-world relevance of our approach. The AIH metric is particularly well-suited for policy-making and risk management, where only ordinal assessments are available, and it enables more informed prioritization of mitigation strategies.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Vei, S., Giudici, P., Sermpezis, P., Vakali, A. I., & Bernardelli, A. E. (2026). AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework. Artificial Intelligence, 358, 104587. https://doi.org/10.1016/j.artint.2026.104587

MLA 9

Vei, Sofia, et al. “AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework.” Artificial Intelligence, vol. 358, 2026, p. 104587. https://doi.org/10.1016/j.artint.2026.104587.

Chicago

Vei, Sofia, Paolo Giudici, Pavlos Sermpezis, Athena I. Vakali, and Adelaide Emma Bernardelli. 2026. “AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework.” Artificial Intelligence 358: 104587. https://doi.org/10.1016/j.artint.2026.104587.

Harvard

Vei, S. et al. (2026) ‘AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework’, Artificial Intelligence, 358, p. 104587. doi:10.1016/j.artint.2026.104587.

IEEE

S. Vei, P. Giudici, P. Sermpezis, A. I. Vakali, and A. E. Bernardelli, “AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework,” Artificial Intelligence, vol. 358, p. 104587, 2026, doi: 10.1016/j.artint.2026.104587.

BibTeX

@article{vei2026harmonics,
  title     = {{AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework}},
  author    = {Vei, Sofia and Giudici, Paolo and Sermpezis, Pavlos and Vakali, Athena I. and Bernardelli, Adelaide Emma},
  journal   = {Artificial Intelligence},
  year      = {2026},
  volume    = {358},
  pages     = {104587},
  doi       = {10.1016/j.artint.2026.104587},
  url       = {https://doi.org/10.1016/j.artint.2026.104587}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI

  • AI risk assessment
  • Gini coefficient
  • inequality theory
  • criticality index
  • personal injury
  • public trust in AI
  • AI risk management

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Vei, S., Giudici, P., Sermpezis, P., Vakali, A. I., & Bernardelli, A. E. (2026). AI Harmonics: A human-centric and harms severity-adaptive AI risk assessment framework. Artificial Intelligence, 358, 104587. https://doi.org/10.1016/j.artint.2026.104587

How many times has this paper been cited?

4 times, according to OpenAlex on 18 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Artificial Intelligence: sign in and open the full text from this page.

Who wrote this paper?

Sofia Vei, Paolo Giudici, Pavlos Sermpezis, Athena I. Vakali, Adelaide Emma Bernardelli.

What is the DOI of this paper?

10.1016/j.artint.2026.104587, permanently at https://doi.org/10.1016/j.artint.2026.104587. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 18 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.