New research indicates that people often associate artificial intelligence systems with a White racial identity, even when the AI lacks any human-like characteristics. This tendency was observed across multiple studies and was found to correlate with increased trust in AI recommendations. Furthermore, AI systems described as being trained on data from Black and Latine adults were assigned less demanding tasks compared to those trained on data from White or unspecified demographics. AI
IMPACT This research highlights potential biases in human perception of AI, which could influence user trust and the deployment of AI in various applications.
RANK_REASON The cluster reports on findings from a published academic study.
Read on Mastodon — fosstodon.org →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →