Subscriber Benefit
As a subscriber you can listen to articles at work, in the car, or while you work out. Subscribe Now
Trial lawyers need tools that help us understand the people who will decide our case. And when it comes to using AI for jury selection, it’s the questions you ask.
A good AI query doesn’t happen by accident. It’s built. And when you build it well, you get analysis that feels less like a generic summary and more like the kind of insight you’d get from a seasoned trial consultant who’s been watching jurors for decades.
At its core, a strong prompt has three parts:
• Intent: What you want the AI to do: analyze, compare, summarize, map, predict.
• Context: The jurisdiction, case type, key facts, statutes and any documents that matter.
• Instruction: The specific task and the format you want the answer in.
Below are several examples of how trial lawyers can frame prompts to surface the juror-level insights on matters that shape deliberations. These aren’t rigid templates. They’re jumping-off points you can adapt to your case theory after sharing your narration of the case facts with AI. Note that I only focused on the first element — intent — as the other two elements are fact-driven based on the specifics of your case.
1. Psychological pressure points in the case
Based on the core narrative of this case, identify the psychological pressure points most likely to influence juror reactions. Explain which personal histories, worldviews or lived experiences would make a juror more receptive or more resistant to our position.
2. Constructing the optimal juror blueprint
Create a comprehensive juror blueprint tailored to this case. Include preferred educational background, work history, social values, political tendencies, media consumption patterns, institutional trust level and any prior exposure to litigation that may shape decision-making.
3. Indicators of hidden bias during voir dire
Describe the behavioral cues, speech patterns and interpersonal signals that may reveal concealed bias during voir dire. Include examples of concerning phrasing, inconsistent responses, avoidance behaviors and nonverbal indicators that suggest predisposition.
4. Surfacing deep-seated or unconscious bias
Develop a set of subtle, open-ended voir dire questions designed to uncover unconscious attitudes related to race, gender, age, authority, corporate conduct, the medical field, insurance companies or law enforcement tailored to the specific themes of this case.
5. Mapping juror personality types
Identify eight to 10 juror personality types commonly encountered in deliberations. For each type, describe hallmark traits, how to recognize them during voir dire and whether they are strategically advantageous or harmful for this case.
6. Strategic voir dire questioning framework
Draft 10 open-ended voir dire questions that encourage jurors to reveal their values, assumptions and emotional orientation toward the case themes. Ensure the questions expose bias and measure narrative receptivity without signaling strategy to opposing counsel.
7. Predictors of juror damages awards
Outline the personal characteristics, life experiences and value systems that tend to correlate with generous awards versus conservative ones. Explain how these factors may manifest during voir dire.
Closing thoughts
Effective and precise input provides indispensable instructions to the algorithm. A well-crafted request can make the difference between productive AI interaction and a confusing one. A good prompt is about understanding the logic and functionality of AI and then using that understanding to achieve the best possible outcome.
AI models need instruction, because they understand context the same way people do. Remember that AI requests are akin to giving a young associate a research project. Whereas humans can intuitively understand the context and nuances of requests, AI models need clearer instructions.
Query development is more than just entering questions; it is a mixture of creativity, critical thinking and an understanding of the capabilities and limits of AI. It’s about understanding the underlying logic and functionality of AI. Effective input serves as the critical mechanism that enables more targeted searches and measurably better results.•
__________
Nordstrom, Ph.D., J.D., works as a trial psychologist primarily in Illinois and Indiana. He can be reached at [email protected]. Opinions expressed are those of the author.
Please enable JavaScript to view this content.