PulseAugur
EN
LIVE 20:32:40

AI-generated faces are becoming more convincing, study finds

By PulseAugur Editorial · [1 sources] ·

A new study published in the Journal of Vision indicates that AI-generated faces are increasingly convincing and difficult to distinguish from real ones. Researchers found that while faces generated by Diffusion Models (DM) were slightly easier to identify as synthetic compared to those from Generative Adversarial Networks (GANs), a majority of AI-generated faces were sufficient to deceive most observers. This growing trustworthiness of synthetic faces has implications for various fields, including technology and psychology. AI

IMPACT AI-generated faces are becoming more convincing, potentially impacting fields like digital forensics, media, and online identity verification.

RANK_REASON Research paper published in an academic journal. [lever_c_demoted from research: ic=1 ai=1.0]

Read on Mastodon — fosstodon.org →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

AI-generated faces are becoming more convincing, study finds

COVERAGE [1]

  1. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    AI-generated faces are becoming more trustworthy "Although DM faces were easier to identify as synthetic than GAN faces, most AI-generated faces were sufficient

    AI-generated faces are becoming more trustworthy "Although DM faces were easier to identify as synthetic than GAN faces, most AI-generated faces were sufficient to fool most people much of the time." Alexis A. McGuire, Maty Bohacek, Hany Farid, Paul J. Taylor, Sophie J. Nightinga…

RELATED TOPICS