PulseAugur
EN
LIVE 02:27:44
中文(ZH) https://www. techbang.com/posts/133671-gpt- 5-research-ai-facial-bias-worsens 【AI也會以貌取人?研究驚爆GPT-5偏見比人類還深,連猴子長相都不放過】 這要看餵了什麼資料訓練 AI 吧!如果訓練資料的人類長相與其行為統計比率 (例如罪犯的照

GPT-5 exhibits facial bias, study finds

A recent study suggests that GPT-5 exhibits facial bias, potentially worse than human bias, even extending to the appearance of monkeys. Researchers indicate that the AI's biases are likely a direct reflection of the data it was trained on, including any statistical correlations between appearance and behavior present in the training datasets. AI

IMPACT Highlights potential ethical concerns and the need for careful data curation in future AI development.

RANK_REASON The item discusses a study about a potential AI model (GPT-5) but does not appear to be a direct release or announcement from the AI lab itself, nor is it a primary research paper.

Read on Mastodon — mastodon.social →

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

GPT-5 exhibits facial bias, study finds

How we ranked this

Signal score
3 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Commentary
The item discusses a study about a potential AI model (GPT-5) but does not appear to be a direct release or announcement from the AI lab itself, nor is it a primary research paper.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
safety, model release
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. Mastodon — mastodon.social TIER_1 中文(ZH) · [email protected] ·

    AI also judges by appearance? Research reveals GPT-5 bias is worse than humans, not even sparing monkeys' looks. This depends on the data used to train the AI! If the training data's human appearance and behavior statistical ratios (e.g., criminals' photos

    https://www. techbang.com/posts/133671-gpt- 5-research-ai-facial-bias-worsens 【AI也會以貌取人?研究驚爆GPT-5偏見比人類還深,連猴子長相都不放過】 這要看餵了什麼資料訓練 AI 吧!如果訓練資料的人類長相與其行為統計比率 (例如罪犯的照片與一般人的照片差異) 就是那樣,那麼呈現出來的結果自然就是那樣了。 # AI