Study finds bias in language models against non-binary users

What happens when the technology meant to protect marginalized voices ends up silencing them? Rebecca Dorn, a research assistant at USC Viterbi's Information Sciences Institute (ISI) has uncovered how large language models (LLMs) that are used to moderate online content are failing queer communities by misinterpreting their language.

LLMs display different cultural tendencies when responding to queries in English and Chinese, study finds

techxplore.com - 07/Jul 13:10

Large language models (LLMs), such as the model underpinning the functioning of OpenAI's conversational platform ChatGPT, are now widely used by...

LLMs display different cultural tendencies when responding to queries in English and Chinese, study finds

techxplore.com - 07/Jul 13:10

Large language models (LLMs), such as the model underpinning the functioning of OpenAI's conversational platform ChatGPT, are now widely used by...

AI Displays Racial Bias Evaluating Mental Health Cases

drugs.com - 09/Jul 10:35

WEDNESDAY, July 9, 2025 — AI programs can exhibit racial bias when evaluating patients for mental health problems, a new study says.Psychiatric...

Test-time training could lead to LLMs that are better at complex reasoning

techxplore.com - 08/Jul 16:20

For all their impressive capabilities, large language models (LLMs) often fall short when given challenging new tasks that require complex reasoning...

Researchers Bypass Meta’s Llama Firewall Using Prompt Injection Vulnerabilities

itsecuritynews.info - 12/Jul 22:31

Researchers at Trendyol, a leading e-commerce platform, have uncovered multiple vulnerabilities in Meta’s Llama Firewall, a suite of tools designed...

Marginalized Americans are highly skeptical of artificial intelligence, research finds

wn.com - 02/Jul 23:10

Artificial intelligence may be...

Marginalized Americans are highly skeptical of artificial intelligence, research finds

wn.com - 02/Jul 23:10

Artificial intelligence may be...

AI chatbots often distort nations' human rights records, study finds

mashable.com - 03/Jul 09:00

LLM bias has the potential to warp perceptions on how nations treat their citizens. LLMs — the data models powering your favorite AI chatbots —...

The OWASP Top 10 for LLM Applications: An Overview of AI Security Risks

itsecuritynews.info - 03/Jul 18:07

The world of AI, especially with Large Language Models (LLMs) and Generative AI, is changing the game. It’s like we’ve unlocked a superpower for...

Test-time training could lead to LLMs that are better at complex reasoning

lovablevibes.co - 08/Jul 16:20

Example of ARC and BBH tasks that the model successfully solves only after applying test-time training. Credit: arXiv (2024). DOI:...

Rubriques :

Study finds bias in language models against non-binary users

Articles similaires

LLMs display different cultural tendencies when responding to queries in English and Chinese, study finds

LLMs display different cultural tendencies when responding to queries in English and Chinese, study finds

AI Displays Racial Bias Evaluating Mental Health Cases

Test-time training could lead to LLMs that are better at complex reasoning

Researchers Bypass Meta’s Llama Firewall Using Prompt Injection Vulnerabilities

Marginalized Americans are highly skeptical of artificial intelligence, research finds

Marginalized Americans are highly skeptical of artificial intelligence, research finds

AI chatbots often distort nations' human rights records, study finds

The OWASP Top 10 for LLM Applications: An Overview of AI Security Risks

Test-time training could lead to LLMs that are better at complex reasoning

Les derniers communiqués