Human detection of political speech deepfakes across transcripts, audio, and video

63 citations

Abstract

Recent advances in technology for hyper-realistic visual and audio effects provoke the concern that deepfake videos of political speeches will soon be indistinguishable from authentic video. We conduct 5 pre-registered randomized experiments with N = 2215 participants to evaluate how accurately humans distinguish real political speeches from fabrications across base rates of misinformation, audio sources, question framings with and without priming, and media modalities. We do not find base rates of misinformation have statistically significant effects on discernment. We find deepfakes with audio produced by the state-of-the-art text-to-speech algorithms are harder to discern than the same deepfakes with voice actor audio. Moreover across all experiments and question framings, we find audio and visual information enables more accurate discernment than text alone: human discernment relies more on how something is said, the audio-visual cues, than what is said, the speech content.

63
Citations
Research
Paper Only

Study specs

Five pre-registered randomized experiments with varied base rates of misinformation, audio sources, question framings, and media modalities were conducted.

Sample Size
N=2,215
Study Type
Experimental Study
Year
2025
Human Data Platform
Prolific

Measured Outcomes

Human accuracy in discerning real political speeches from deepfakes across media formats and contextual variables.

Peer Review & Critical Discussion

3 threads

Potential Selection Bias in 2023 Cohort

DSJDr. Sarah J.
Verified PhD Candidate
12 replies

The participant pool shows a concerning overrepresentation of users from high-income demographics. Looking at Table 3, we can see that 78% of respondents had annual incomes above $75k, which significantly limits the generalizability of these findings to broader populations.

2 hours ago

Non-naive Participants Issue

MCM. Chen (OpenAI)
Data Scientist
8 replies

I've noticed a methodological concern regarding participant naivety. Given that Prolific users often complete multiple studies, there's a real risk that participants had prior exposure to similar experimental paradigms, which could confound the results.

5 hours ago

RLHF Applicability to This Study Design

PRWProf. R. Williams
Verified Researcher
15 replies

The implications for RLHF training pipelines are understated. If we accept the authors' conclusions about preference stability, this has direct consequences for how we should structure reward model training. The temporal decay effect described in Section 4.2 is particularly relevant.

1 day ago

Verify your expertise to join discussion

Create an account and verify your credentials to participate in peer discussions.