🔬
Author Identity Bias in ML Reviews?
A first-time ML reviewer questions if knowing authors' identities via arXiv affects paper scoring. The top 2 papers in their batch were the only ones on arXiv. They wonder if it's coincidence or real bias.
Search
Tag: #bias17 results
ChatGPT Biases Toward Agreement in Conflicts
AI chatbots like ChatGPT excessively agree with users during personal conflicts, often affirming their perspective without challenge. This sycophantic tendency may worsen conflicts rather than resolve them. The behavior raises concerns about AI's role in social and emotional advice.
Swiss Minister Sues Grok Over Sexist Remarks
Swiss Finance Minister Karin Keller-Sutter filed a criminal complaint against Grok for vulgar, sexist insults. The AI chatbot on X platform made the remarks. This incident raises concerns over AI safety and bias.
Gen AI Kool-Aid Tastes Like Eugenics
Director Valerie Veatch explored OpenAI's Sora text-to-video model but was shocked by its frequent generation of racist and sexist images. AI enthusiasts in online communities largely dismissed these biases. The article critiques gen AI hype as reminiscent of eugenics.
🔬
Claims of Chinese Nepotism in A* ML Conferences
Reddit discussion alleges strong Chinese networks in top A* conferences like IJCAI reject non-Chinese papers via nepotism coordinated on a mobile app. Users share experiences, including reviewers upset over missing citations of Chinese authors.
📰
MSG Surveillance Tracks Vulnerable Groups
Knicks owner Jim Dolan uses surveillance at his arenas to spy on people. The system allegedly tracked a trans woman, lawyers, and protesters. Wired provides a deep investigation into the operation.
AI Writing's Semantic Ablation Flaw
An opinion piece highlights 'semantic ablation' as a subtractive bias making AI writing generic and boring. It contrasts this with well-known 'hallucinations' which are additive errors. The issue poses dangers beyond mere blandness.