ResearchBuzz: FirehosePsyPost: Scientists shocked to find AI’s social desirability bias “exceeds typical human standards”. “A new study published in PNAS Nexus reveals that large language models, which are advanced artificial intelligence systems, demonstrate a tendency to present themselves in a favorable light when taking personality tests. This ‘social desirability bias’ leads these models to score higher […]
PsyPostScientists shocked to find AI’s social desirability bias “exceeds typical human standards” https://www.psypost.org/scientists-shocked-to-find-ais-social-desirability-bias-exceeds-typical-human-standards/?utm_source=dlvr.it&utm_medium=mastodon #AI #ArtificialIntelligence #SocialDesirabilityBias #PersonalityTests #Research
Scientists shocked to find AI’s social desirability bias “exceeds typical human standards”
Large language models (LLMs) exhibit social desirability bias on personality tests, scoring higher on positive traits and lower on negative ones. This bias is stronger in newer, larger models, potentially impacting their use in research simulating human behavior.