A.I. and Legal Decision Making

Guest Lecture: PSYC401 Psychology & Law

Emma Marshall, JD, MA

University of Nebraska-Lincoln

March 11, 2025

What is Artificial Intelligence?

Understanding of AI-systems

Kennedy, Tyson, and Saks (2023)

Machine Learning & “AI”

Iterative Process

Scientific Method?

AI & Other Cognitive Sciences

Van Rooij et al. (2024)

What do we mean by “AI”?

Van Rooij et al. (2024)

Code & Math (NOT biology)

AI is just a (powerful) TOOL!

AI makes predictions based on pre-existing data

Bender et al. (2021)

When is AI Used to Make Legal Decisions?

…all the time!

(“Inescapable AI: The Ways AI Decides How Low Income People Work, Live, Learn, and Survive” 2024)

Types of Algorithms & “AI”

Algorithm Type	Process Description	Type of Data Used	Legal Examples
Rule-Based Decision Making	If condition fulfilled then activity 1 else activity 2	Boolean data (yes or no)	Court Scheduling
Statistical Reasoning	Simple regression	Numerical data allowing for curve fitting	Risk Assessment
Machine Learning	Classification tasks	Arbitrary data that needs to be abstracted into numbers	Facial recognition
Artificial Intelligence	Dynamic adaptation to novelty	Autonomous selection of best methodology	Intelligent digital assistant

Law means many things…

Law-Psychology does many things…

Determine Facts

• Trademarks
• Obscenity
• Damages

Used to Make Law - Legislative facts (need to regulate harm, prevalence and who is impacted) - Constitutional principles (First Amendment, Due Process, Fourth Amendment, etc.)

Used to Provide (Social) Context

Litigation Strategy Where to file and who to have hear your case?

What can go wrong?

Good at: Document review (TAR)

Bad at: Writing Police Reports?

Ferguson (2024)

Fairness

• Historical data bias
• Proxy variables
  • i.e., zip codes, arrest records
• Definitional challenges
  • What is fair?)

(Starke et al. 2022; Dolata and Schwabe 2024)

COMPASS: Correctional Offender Management Profiling for Alternative Sanctions

Accuracy

• Performance Metrics:
  • Accuracy (overall population)
  • Precision (individual cases)
  • Recall (detection rate)
• Data quality issues
• Training data limitations

How to Train Your Data

Hallucinations!?!?

https://undark.org/2023/04/06/chatgpt-isnt-hallucinating-its-bullshitting/

Accountability

• “Black box” systems
• Proprietary software
• Distributed widely
• Responsibility gaps
• Trust Paradox Kreps et al. (2023)

Transparency

• Propriety algorithms
• Limited public access to:
  • Source code
  • Training data
  • Testing methods
• BUT needed for contestability

Example: AI & Deepfake Pornography

What are Deepfakes?

Technical Background

• Uses AI/machine learning to produce/alter images, audio & videos
• Can create realistic content with minimal source material

How They Work

1. Neural Networks - Uses deep learning algorithms - Trains on facial data - Creates mathematical models of faces 2. GANs (Generative Adversarial Networks) - Generator network creates images - Discriminator network evaluates authenticity - Systems work together to improve output

How are they used?

• Entertainment (e.g., make Harrison Ford look young)
• Political manipulation
• Advertising
• **Pornography

96% of all deepfakes are porn

Problem with Deepfake Porn

• Devastating personal and professional consequences
• Difficulty removing content once posted (copywrite)
• Limited legal recourse

(Suslavich 2023; Mania 2020)

What is new about AI Deepfakes?

(Anslow 2024)

Which Face Is Real?

Pre-existing Criminal Statutes

• Identity theft (limited application)
• Cyberstalking (18 U.S.C. §2261A)
• Federal legislation (NDAA 2020)

Copyright Law

• Generally ineffective for deepfakes (except for DCMA)
• Faces/voices not always copyrightable
• Fair use/transformative work defenses

Right of Publicity

• Limited to commercial use
• Only ~30 states recognize it
• New York’s digital replica protection
• Section 230 immunity challenges

Defamation

Public Figures

• Must prove actual malice
• Higher burden of proof
• First Amendment considerations

Private Individuals

• Lower standard than public figures
• State law variations
• Must show actual harm

Legal Barriers

1. Platform Immunity
   • Section 230 protections
   • Limited ISP liability
2. First Amendment Issues
   • Prior restraint doctrine
   • Free speech protections
   • Parody defenses

Parody Defense

• Hustler Magazine v. Falwell
• “Reasonable person” standard
• Disclaimer effectiveness
• Quality considerations

Practical Issues

• Anonymous creators
• Jurisdiction on internet
• Rapid spread of content & accumulated harm

So what has been done?

Current Legislative Approaches

1. State Laws
   • California (A.B. 730)
   • Virginia (nonconsensual pornography)
   • New York (right of publicity)
2. Federal Proposals
   • DEEP FAKES Accountability Act
   • Watermark requirements
   • Criminal penalties

What role can Law-Psychology play?

Questions?

About

 

• Website
• em.marshall47@gmail.com
• @emmarshall

References

Anslow, Louis. 2024. “The 1912 War on Fake Photos.” Pessimists Archive. https://newsletter.pessimistsarchive.org/p/the-1912-war-on-fake-photos.

Bender, Emily M., Timnit Gebru, Angelina McMillan-Major, and Shmargaret Shmitchell. 2021. “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜.” In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610–23. Virtual Event Canada: ACM. https://doi.org/10.1145/3442188.3445922.

Dolata, Mateusz, and Gerhard Schwabe. 2024. “Towards the Socio-Algorithmic Construction of Fairness: The Case of Automatic Price-Surging in Ride-Hailing.” International Journal of Human–Computer Interaction 40 (1): 55–65. https://doi.org/10.1080/10447318.2023.2210887.

Ferguson, Andrew Guthrie. 2024. “Generative Suspicion and the Risks of AI-Assisted Police Reports.”

“Inescapable AI: The Ways AI Decides How Low Income People Work, Live, Learn, and Survive.” 2024. Techtonic Justice.

Kennedy, Brian, Alec Tyson, and Emily Saks. 2023. “What Americans Know About Everyday Uses of Artificial Intelligence Pew Research Center.” Pew Research. https://www.pewresearch.org/science/2023/02/15/public-awareness-of-artificial-intelligence-in-everyday-activities/.

Kreps, Sarah, Julie George, Paul Lushenko, and Adi Rao. 2023. “Exploring the Artificial Intelligence ‘Trust Paradox’: Evidence from a Survey Experiment in the United States.” Edited by Hans H. Tung. PLOS ONE 18 (7): e0288109. https://doi.org/10.1371/journal.pone.0288109.

Mania, Karolina. 2020. “The Legal Implications and Remedies Concerning Revenge Porn and Fake Porn: A Common Law Perspective.” Sexuality & Culture 24 (6): 2079–97. https://doi.org/10.1007/s12119-020-09738-0.

Starke, Christopher, Janine Baleis, Birte Keller, and Frank Marcinkowski. 2022. “Fairness Perceptions of Algorithmic Decision-Making: A Systematic Review of the Empirical Literature.” Big Data & Society 9 (2): 20539517221115189. https://doi.org/10.1177/20539517221115189.

Suslavich, Benjamin T. 2023. “Nonconsensual Deepfakes: A "Deep Problem" for Victims.” Journal of Science and Technology 33: 160–88. https://towardsdatascience.com/multi-layer-neural-networks-with-sigmoidfunction-deep-learning-for-rookies-2-bf464f09eb7f.

Van Rooij, Iris, Olivia Guest, Federico Adolfi, Ronald De Haan, Antonina Kolokolova, and Patricia Rich. 2024. “Reclaiming AI as a Theoretical Tool for Cognitive Science.” Computational Brain & Behavior 7 (4): 616–36. https://doi.org/10.1007/s42113-024-00217-5.