Child Abuse, AI, and the Forensic Interview

In this episode of 'One in Ten,' host Teresa Huizar speaks with Liisa Järvilehto, a psychologist and Ph.D. candidate at the University of Helsinki, about the positive uses of AI in child abuse investigations and forensic interviews. The conversation addresses the common misuse of AI and explores its potential in assisting professionals by proposing hypotheses, generating question sets, and more. The discussion delves into the application of large language models (LLMs) in generating alternative hypotheses and the nuances of using these tools to avoid confirmation bias in interviews. Huizar and Järvilehto also touch on the practical implications for current practitioners and future research directions. 

 

Time Stamps: 

 

00:00 Introduction to the Episode 

00:00 Introduction to the Episode 

00:22 Exploring AI in Child Abuse Investigations 

01:06 Introducing Liisa Järvilehto and Her Research 

01:48 Challenges in Child Abuse Investigations 

04:24 The Role of Large Language Models 

06:28 Addressing Bias in Investigations 

09:13 Hypothesis Testing in Forensic Interviews 

12:18 Study Design and Findings 

25:54 Implications for Practitioners 

33:41 Future Research Directions 

36:49 Conclusion and Final Thoughts 

Resources:

Pre-interview hypothesis generation: large language models (LLMs) show promise for child abuse investigations

Support the show

Did you like this episode? Please leave us a review on Apple Podcasts.