Can AI tell when you’re lying?

A new study is diving deeper into how well artificial intelligence (AI) can understand humans by using it to detect human deception.

Published in the *Journal of Communication*, the research involved Michigan State University and University of Oklahoma researchers conducting 12 experiments with over 19,000 AI participants. The goal was to examine how well AI personas could detect deception and truth from human subjects.

“This research aims to understand how well AI can aid in deception detection and simulate human data in social scientific research, as well as caution professionals when using large language models for lie detection,” says David Markowitz, associate professor of communication in the MSU College of Communication Arts and Sciences and lead author of the study.

### Comparing AI and Human Deception Detection

To evaluate AI’s performance relative to humans, the researchers drew on Truth-Default Theory (TDT). TDT suggests that people are mostly honest most of the time, and we naturally tend to believe others are telling the truth.

“Humans have a natural truth bias — we generally assume others are being honest, regardless of whether they actually are,” Markowitz explains. “This tendency is thought to be evolutionarily useful, since constantly doubting everyone would take much effort, make everyday life difficult, and strain relationships.”

This theory helped the team compare how AI behaves in similar situations to human participants.

### How the Study Was Conducted

The researchers used the Viewpoints AI research platform to present audiovisual or audio-only media of humans for various AI personas to judge. These AI judges were tasked with determining if the human subjects were lying or telling the truth—and providing a rationale for their decisions.

Several variables were tested, including:
– Media type (audiovisual or audio-only)
– Contextual background (information or circumstances explaining why something happens)
– Lie-truth base-rates (proportions of honest and deceptive communication)
– Persona of the AI (identities designed to act and speak like real people)

The goal was to assess how these factors affected AI’s deception detection accuracy.

### Key Findings

One study found that AI exhibited a lie bias—it was far more accurate at detecting lies (85.8%) compared to truths (19.5%). In short interrogation settings, AI’s deception accuracy was comparable to that of humans.

However, in non-interrogation contexts, such as evaluating statements about friends, AI displayed a truth bias, aligning more closely with human performance.

Overall, the results showed that AI tends to be more lie-biased and is generally much less accurate than humans at deception detection.

“Our main goal was to see what we could learn about AI by including it as a participant in deception detection experiments,” Markowitz says. “In this study, and with the model we used, AI turned out to be sensitive to context—but that didn’t make it better at spotting lies.”

### Implications and Future Directions

The study’s findings suggest that AI’s deception detection results do not match human accuracy. This indicates that “humanness” might be a critical limit or boundary condition for how deception detection theories apply.

While the use of AI for deception detection may seem unbiased and technologically advanced, the research highlights that significant progress is needed before generative AI can be reliably used in this domain.

“It’s easy to see why people might want to use AI to spot lies—it seems like a high-tech, potentially fair, and possibly unbiased solution. But our research shows that we’re not there yet,” Markowitz concludes. “Both researchers and professionals need to make major improvements before AI can truly handle deception detection.”

—

This study sheds important light on the current capabilities and limitations of AI in understanding and interpreting human honesty, signaling caution for its practical use in lie detection at this time.
https://www.futurity.org/can-ai-tell-when-youre-lying-3304212/

You may also like

The Walking Dead’s Zombie Virus Origin Explained

2025 年 11 月 10 日

XPL Rallies After Plasma Reveals Collaboration with Daylight Energy

2025 年 11 月 10 日

Razer DeathAdder V4 Pro mouse review: Serious bite

2025 年 11 月 10 日