Machine Learning Models in Parole Decision-Making: A Pilot Study Comparing AI and Human Judgment

Abstract

This pilot study evaluates three machine learning models—Claude, ChatGPT, and Gemini—in parole decision-making by comparing their recommendations against human judicial decisions. Analyzing 150 cases across sexual, drug-related, and violent offenses, the study demonstrates alignment rates of 56%–68% between artificial intelligence (AI)-generated and human decisions. Findings reveal a consistent rehabilitative bias in AI recommendations, with statistical analyses indicating robust patterns across offense categories. This research establishes an empirical foundation for understanding AI’s potential as a decision-support tool in parole processes while preserving essential human judgment. While focusing on Israel, the findings offer a preliminary basis for considering similar AI integrations across different jurisdictions. These results underscore the importance of developing clear criminal justice policies and ethical frameworks to guide the responsible use of AI in parole decisions, ensuring that technological tools support rather than replace human judgment and enhance public trust in the system.

Keywords

artificial intelligence parole decision-making criminal justice predictive modeling ethical AI machine-human collaboration early release

Get full access to this article

View all access options for this article.

References

Ajunwa

(2020). The paradox of automation as an anti-bias intervention. Cardozo Law Review, 41(5), 1671–1742.

Alarie

Niblett

Yoon

A. H.

(2016). Law in the future. University of Toronto Law Journal, 66(4), 423–428.

Angwin

Larson

Mattu

Kirchner

(2016, May 23). Machine bias: There is software used across the country to predict future criminals, and it is biased against blacks. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing

Anthropic. (2025). Introducing Claude 3.5 Sonnet. https://www.anthropic.com/news/claude-3-5-sonnet

Arowosegbe

J. O.

(2023). Data bias, intelligent systems, and criminal justice outcomes. International Journal of Law and Information Technology, 31(1), 22–45.

Barocas

Selbst

A. D.

(2016). Big data’s disparate impact. California Law Review, 104(3), 671–732. https://doi.org/10.15779/Z38BG31

Berk

Heidari

Jabbari

Kearns

Roth

(2021). Fairness in criminal justice risk assessments: The state of the art. Sociological Methods & Research, 50(1), 3–44. https://doi.org/10.1177/0049124118782533

Binns

(2018). Fairness in machine learning: Lessons from political philosophy. Proceedings of Machine Learning Research, 81, 149–159.

Dagan

(2023). Parole as resentencing: Exploring the punitive accounts of parole decision-making through the comparative case study of Israel. European Journal of Criminology, 20(4), 1231–1250. https://doi.org/10.1177/14773708211039635

10.

Dagan

Timor

Ronel

(2013). Parole release objectives as viewed by parole board heads and their discretion in their application. Tzohar LeBeit HaSohar, 15, 62–81.

11.

Dou

(2025). Towards just AI: Challenges and solution framework for algorithmic discrimination in judicial systems. International Journal of Digital Law and Governance. Advance online publication. https://doi.org/10.1515/ijdlg-2024-0020

12.

Elo

Kyngäs

(2008). The qualitative content analysis process. Journal of Advanced Nursing, 62(1), 107–115.

13.

Farayola

M. M.

Tal

Bendechache

Saber

Connolly

(2023). Fairness of AI in predicting the risk of recidivism: Review and phase mapping of AI fairness techniques. In Proceedings of the 18th International Conference on Availability, Reliability and Security (ARES 2023) (pp. 1–10). ACM. https://doi.org/10.1145/3600160.3605033

14.

Farber

(2024). AI in terrorism sentencing: Evaluating predictive accuracy and ethical implications. Criminal Justice Studies, 37(4), 352–370. https://doi.org/10.1080/1478601X.2024.2415558

15.

Farber

(2025). AI as a decision support tool in forensic image analysis: A pilot study on integrating large language models into crime scene investigation workflows. Journal of Forensic Sciences. Advance online publication. https://doi.org/10.1111/1556-4029.70035

16.

Ganguli

Hernandez

Lovitt

Askell

Bai

Chen

Conerly

Dassarma

Drain

Elhage

El Showk

Fort

Hatfield-Dodds

Henighan

Johnston

Jones

Joseph

Kernian

Kravec

. . .Clark

(2022, June). Predictability and surprise in large generative models. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (pp. 1747–1764). ACM.

17.

Google. (2025). Gemini Pro. https://deepmind.google/technologies/gemini/pro

18.

Harris

(2022). Achieving consistency in sentencing. Oxford University Press.

19.

Hassija

Chamola

Mahapatra

Singal

Goel

Huang

Scardapane

Spinelli

Mahmud

Hussain

(2024). Interpreting black-box models: A review on explainable artificial intelligence. Cognitive Computation, 16, 45–74. https://doi.org/10.1007/s12559-023-10179-8

20.

Huq

A. Z.

(2019). Racial equity in algorithmic criminal justice. Duke Law Journal, 68(6), 1043–1134.

21.

Israel Ministry of Justice. (2012). Prisoner parole regulations. https://www.justice.gov.il/Regulations

22.

Javed

(2025). Bias in adjudication: Investigating the impact of artificial intelligence, media, financial, and legal institutions in pursuit of social justice. PLOS ONE, 20(1), Article e0315270.

23.

Jiang

Yang

(2024). Agents on the Bench: Large language model-based multi-agent framework for trustworthy digital justice. arXiv. https://arxiv.org/abs/2412.18697

24.

Katz

D. M.

Bommarito

M. J.

Blackman

(2017). A general approach for predicting the behavior of the United States Supreme Court. PLOS ONE, 12(4), Article e0174698.

25.

Kleinberg

Lakkaraju

Leskovec

Ludwig

Mullainathan

(2018). Human decisions and machine predictions. The Quarterly Journal of Economics, 133(1), 237–293. https://doi.org/10.1093/qje/qjx032

26.

Krippendorff

(2018). Content analysis: An introduction to its methodology (4th ed.). Sage.

27.

Landau

Dempsey

J. X.

Kamar

Bellovin

S. M.

Pool

(2024). Challenging the machine: Contestability in government AI systems. arXiv preprint arXiv:2406.10430

28.

Laqueur

H. S.

Copus

R. W.

(2024). An algorithmic assessment of parole decisions. Journal of Quantitative Criminology, 40(1), 151–188.

29.

Lincoln

Y. S.

Guba

E. G.

(1985). Naturalistic inquiry. Sage.

30.

March

J. G.

Simon

H. A.

(1993). Organizations. Wiley-Blackwell.

31.

Mehrabi

Morstatter

Saxena

Lerman

Galstyan

(2021). A survey on bias and fairness in machine learning. ACM Computing Surveys, 54(6), Article 115. https://doi.org/10.1145/3457607

32.

O’Neil

(2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Crown. https://doi.org/10.5860/crl.78.3.403

33.

OpenAI. (2025). GPT-4 Turbo in the OpenAI API. https://help.openai.com/en/articles/8555510-gpt-4-turbo-in-the-openai-api

34.

R. M.

Solow-Niederman

(2019). Developing artificially intelligent justice. Stanford Technology Law Review, 22(2), 242–289.

35.

Ryberg

(2024). Criminal justice and artificial intelligence: How should we assess the performance of sentencing algorithms? Philosophy & Technology, 37(1), 9.

36.

Ryberg

(2025). Artificial intelligence and criminal justice: How to use algorithmic sentencing support in real-life (and ethically non-ideal) penal systems? AI and Ethics. Advance online publication. https://doi.org/10.1007/s43681-024-00655-8

37.

Starr

S. B.

(2014). Evidence-based sentencing and the scientific rationalization of discrimination. Stanford Law Review, 66, 803–872.

38.

Stevenson

M. T.

Mayson

S. G.

(2017). Bail reform: New directions for pretrial detention and release. https://core.ac.uk/works/49697666/?source=1&algorithmId=15&similarToDoc=68767181&similarToDocKey=CORE&recSetID=60147e29-09cc-482f-9fb1-f6ff6a8604e1&position=4&recommendation_type=same_repo&otherRecs=40165153%2C66933836%2C85635782%2C49697666%2C120185661

39.

Taylor

(2023). Justice by algorithm: The limits of AI in criminal sentencing. Criminal Justice Ethics, 42(3), 193–213.

40.

U.S. Department of Justice, Office of Legal Policy. (2024). Artificial intelligence and criminal justice: Final report. U.S. Government Publishing Office.