Artificial Intelligence (AI) Trust Framework and Maturity Model: Applying an Entropy Lens to Improve Security, Privacy, and Ethical AI

Michael Mylrea; Nikki Robinson

doi:10.3390/e25101429

journal article Open Access Oct 09, 2023

Artificial Intelligence (AI) Trust Framework and Maturity Model: Applying an Entropy Lens to Improve Security, Privacy, and Ethical AI

Michael Mylrea Nikki Robinson

Entropy Vol. 25 No. 10 pp. 1429 · MDPI AG

View at Publisher Save 10.3390/e25101429

Abstract

Recent advancements in artificial intelligence (AI) technology have raised concerns about the ethical, moral, and legal safeguards. There is a pressing need to improve metrics for assessing security and privacy of AI systems and to manage AI technology in a more ethical manner. To address these challenges, an AI Trust Framework and Maturity Model is proposed to enhance trust in the design and management of AI systems. Trust in AI involves an agreed-upon understanding between humans and machines about system performance. The framework utilizes an “entropy lens” to root the study in information theory and enhance transparency and trust in “black box” AI systems, which lack ethical guardrails. High entropy in AI systems can decrease human trust, particularly in uncertain and competitive environments. The research draws inspiration from entropy studies to improve trust and performance in autonomous human–machine teams and systems, including interconnected elements in hierarchical systems. Applying this lens to improve trust in AI also highlights new opportunities to optimize performance in teams. Two use cases are described to validate the AI framework’s ability to measure trust in the design and management of AI systems.

Topics

No keywords indexed for this article. Browse by subject →

References

97

[1]

Pattee, H.H. (1976). Hierarchy Theory: The Challenge of Complex Systems, George Braziller.

[2]

Conant "Laws of information which govern systems" IEEE Trans. Syst. Man Cybern. (1976) 10.1109/tsmc.1976.5408775

[3]

Lawless, W.F. (2019). The interdependence of autonomous human-machine teams: The entropy of teams, but not individuals, advances science. Entropy, 21. 10.3390/e21121195

[4]

(2023, July 21). Insects, Biological Collectives. Available online: https://ssr.seas.harvard.edu/insect-collectives.

[5]

(2023, July 21). Mother Tree Project. Available online: https://mothertreeproject.org.

[6]

Qi "Trustworthy AI: From Principles to Practices" ACM Comput. Surv. (2023) 10.1145/3479428

[7]

Hu, W., and Tan, Y. (2023). Data Mining and Big Data: 7th International Conference, DMBD 2022, Beijing, China, 21–24 November 2022, Proceedings, Part II, Springer.

[8]

A Brief History of Artificial Intelligence: On the Past, Present, and Future of Artificial Intelligence

Michael Haenlein, Andreas Kaplan

California Management Review 2019 10.1177/0008125619864925

[9]

Hoffman "A study in cognitive entrenchment: Negative entropy or negative complexity?" J. Cogn. Eng. Decis. Mak. (2018)

[10]

Mann "Collective decision making by rational individuals" Proc. Natl. Acad. Sci. USA (2018) 10.1073/pnas.1811964115

[11]

Bubeck, S., Chandrasekaran, V., Eldan, R., Gehrke, J., Horvitz, E., Kamar, E., Lee, P., Lee, Y., Li, Y., and Zhang, Y. (2023). Sparks of artificial general intelligence: Early experiments with gpt-4. arXiv.

[12]

Gierlichs, B., Batina, L., Tuyls, P., and Preneel, B. (2008). International Workshop on Cryptographic Hardware and Embedded Systems, Springer.

[13]

Wang, S., Li, Y., Zhao, A., and Wang, Q. (2021, January 23–25). Privacy Protection in Federated Learning Based on Differential Privacy and Mutual Information. Proceedings of the 2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture 2021, Manchester, UK. 10.1145/3495018.3495093

[14]

Joyce, J.M. (2011). International Encyclopedia of Statistical Science, Springer.

[15]

Shin "Channel capacity analysis of non-orthogonal multiple access with OAM-MIMO system" IEEE Wirel. Commun. Lett. (2020) 10.1109/lwc.2020.2994355

[16]

Chuprov, S., Viksnin, I., Kim, I., Reznikand, L., and Khokhlov, I. (August, January 1). Reputation and Trust Models with Data Quality Metrics for Improving Autonomous Vehicles Traffic Security and Safety. Proceedings of the 2020 IEEE Systems Security Symposium (SSS), Crystal City, VA, USA. 10.1109/sss47320.2020.9174269

[17]

Mylrea, M. (2023). Putting AI in the Critical Loop Assured Trust and Autonomy in Human-Machine Teams, Elsevier.

[18]

Pandolfo, C. (2023, July 21). ‘Godfather of Artificial Intelligence’ Says AI Is Close to Being Smarter than Us, Could End Humanity. Fox Busines. Available online: https://www.foxbusiness.com/technology/godfather-artificial-intelligence-says-ai-close-being-smarter-could-end-humanity.

[19]

Chen "Challenges and corresponding solutions of generative adversarial networks (GANs): A survey study" J. Phys. Conf. Ser. (2021) 10.1088/1742-6596/1827/1/012066

[20]

Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI

Alejandro Barredo Arrieta, Natalia Díaz-Rodríguez, Javier Del Ser et al.

Information Fusion 2020 10.1016/j.inffus.2019.12.012

[21]

A Survey of Methods for Explaining Black Box Models

Riccardo Guidotti, Anna Monreale, Salvatore Ruggieri et al.

ACM Computing Surveys 2018 10.1145/3236009

[22]

Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead

Cynthia Rudin

Nature Machine Intelligence 2019 10.1038/s42256-019-0048-x

[23]

Zafar "Deterministic local interpretable model-agnostic explanations for stable explainability" Mach. Learn. Knowl. Extr. (2021) 10.3390/make3030027

[24]

Nohara, Y., Matsumoto, K., Soejima, H., and Nakashima, N. (2019, January 7–10). Explanation of machine learning models using improved Shapley Additive Explanation. Proceedings of the 10th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics, Niagara Falls, NY, USA. 10.1145/3307339.3343255

[25]

A Survey of Contrastive and Counterfactual Explanation Generation Methods for Explainable Artificial Intelligence

Ilia Stepin, Jose M. Alonso, Alejandro Catala et al.

IEEE Access 2021 10.1109/access.2021.3051315

[26]

Rosenfeld "Explainability in human-agent systems" Auton. Agents Multi Agent Syst. (2019) 10.1007/s10458-019-09408-y

[27]

Rosenfeld, A. (2021, January 3–7). Better Metrics for Evaluating Explainable Artificial Intelligence: Blue Sky Ideas Track. Proceedings of the 21th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2021), Online.

[28]

Cavoukian "Privacy by Design in the Age of Artificial Intelligence" Priv. Data Prot. J. (2019)

[29]

Obar "The biggest lie on the internet: Ignoring the privacy policies and terms of service policies of social networking services" Inf. Commun. Soc. (2018)

[30]

Solove "Privacy self-management and the consent dilemma" Harv. Law Rev. (2013)

[31]

Shadbolt "Transparent and explainable AI for preserving privacy" IEEE Internet Comput. (2017)

[32]

European Commission (2018). Guidelines on the Application of the European Union General Data Protection Regulation, European Commission.

[33]

Cavoukian "Privacy by Design: The 7 Foundational Principles" Inf. Priv. Comm. Ont. Can. (2009)

[34]

Narayanan, A., and Shmatikov, V. (2008, January 18–22). Robust de-anonymization of large sparse datasets. Proceedings of the 2008 IEEE Symposium on Security and Privacy, Oakland, CA, USA. 10.1109/sp.2008.33

[35]

El Emam, K., and Arbuckle, L. (2014). Anonymizing Health Data: Case Studies and Methods to Get You Started, O’Reilly Media, Inc.

[36]

European Union Agency for Cybersecurity (2020). Cybersecurity Certification Scheme for cloud SERVICES, ENISA.

[37]

Cavoukian, A. (2011). Privacy by Design: The Definitive Workshop, Springer. 10.4018/978-1-61350-501-4.ch007

[38]

Gupta, J.N., and Sharma, S. (2009). Handbook of Research on Information Security and Assurance, IGI Global. 10.4018/978-1-59904-855-0

[39]

Kelly, J. (2023, July 25). Artificial Intelligence Is Already Causing Disruption and Job Loss at IBM And Chegg. Forbes, 2 May 2023. Available online: https://www.forbes.com/sites/jackkelly/2023/05/02/artificial-intelligence-is-already-causing-disruption-and-job-losses-at-ibm-and-chegg/?sh=2a1dc61975a4.

[40]

OpenAI (2023, May 10). ChatGPT. Available online: https://openai.com.

[41]

National Institute of Standards and Technology (2023, July 21). NIST AI Metrics Suite, Available online: https://pages.nist.gov/ai-metrics/.

[42]

Caruana, R., Lou, Y., Gehrke, J., Koch, P., Sturm, M., and Elhadad, N. (2015, January 10–13). Intelligible models for healthcare: Predicting pneumonia risk and hospital 30-day readmission. Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Sydney, NSW, Australia. 10.1145/2783258.2788613

[43]

Doshi-Velez, F., Kortz, M., Budish, R., Bavitz, C., Gershman, S., O’Brien, D., Scott, K., Schieber, S., Waldo, J., and Weinberger, D. (2019). Accountability of AI under the law: The role of explanation. arXiv.

[44]

Hendrycks, D., and Gimpel, K. (2016). A baseline for detecting misclassified and out-of-distribution examples in neural networks. arXiv.

[45]

Geirhos, R., Rubisch, P., Michaelis, C., Bethge, M., Wichmann, F.A., and Brendel, W. (2018). Imagenet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness. arXiv.

[46]

Liang, S., Li, Y., and Srikant, R. (2018). Enhancing the reliability of out-of-distribution image detection in neural networks. arXiv.

[47]

Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., and Fergus, R. (2013). Intriguing properties of neural networks. arXiv.

[48]

Zhang, C., Bengio, S., Hardt, M., Recht, B., and Vinyals, O. (2019). Understanding deep learning requires rethinking generalization. arXiv.

[49]

Ma, L., Juefei-Xu, F., Zhang, F., Sun, J., Xue, M., Li, B., Chen, C., Su, T., Li, L., and Liu, Y. (2018, January 3–7). DeepGauge: Multi-granularity testing criteria for deep learning systems. Proceedings of the 2018 ACM Joint Meeting on European Software Engineering Conference and Symposium on the Foundations of Software Engineering, Montpellier, France. 10.1145/3238147.3238202

[50]

Ala-Pietilä, P., Bonnet, Y., Bergmann, U., Bielikova, M., Bonefeld-Dahl, C., Bauer, W., Bouarfa, L., Chatila, R., Coeckelbergh, M., and Dignum, V. (2020). The Assessment List for Trustworthy Artificial Intelligence (ALTAI), European Commission.

Showing 50 of 97 references

Metrics

75

Citations

97

References

Details

Published: Oct 09, 2023
Vol/Issue: 25(10)
Pages: 1429
License: View

Authors

M

Michael Mylrea

Department of Computer Science & Engineering, Institute of Data Science and Computing, University of Miami, Coral Gables, FL 33146, USA

N

Nikki Robinson

Department of Computer and Data Science, Capitol Technology University, Laurel, ME 20708, USA

Cite This Article

Michael Mylrea, Nikki Robinson (2023). Artificial Intelligence (AI) Trust Framework and Maturity Model: Applying an Entropy Lens to Improve Security, Privacy, and Ethical AI. Entropy, 25(10), 1429. https://doi.org/10.3390/e25101429

Artificial Intelligence (AI) Trust Framework and Maturity Model: Applying an Entropy Lens to Improve Security, Privacy, and Ethical AI

You May Also Like