Skip to content
GCC AI Research

Search

Results for "human rights"

Enabling Practical and Rich User Digitization

MBZUAI ·

A computer science vision involves computing devices becoming proactive assistants, enhancing various aspects of life through user digitization. Current devices provide coarse digital representations of users, but there's significant potential for improvement. Karan, a Ph.D. candidate at CMU, develops technologies for consumer devices to capture richer user representations without sacrificing practicality. Why it matters: Advancements in user digitization can lead to improved extended reality experiences, health tracking, and more productive work environments, enhancing the utility of consumer devices.

Project NATHR-G1 - The Story behind the Innovative Solution for a more Humane World

TII ·

A team led by the Technology Innovation Institute (TII) in Abu Dhabi has developed NATHR-G1, a ground penetrating radar for detecting landmines and unexploded ordnance. The project, involving researchers from Colombia, Germany, Sweden, and Switzerland, builds on earlier work using radar to detect buried objects. NATHR-G1 incorporates machine learning for advanced signal processing and object identification. Why it matters: This humanitarian application of AI and robotics based in the UAE could significantly reduce casualties from landmines and other explosive remnants of war.

Profiling News Media for Factuality and Bias Using LLMs and the Fact-Checking Methodology of Human Experts

arXiv ·

A new methodology emulating fact-checker criteria assesses news outlet factuality and bias using LLMs. The approach uses prompts based on fact-checking criteria to elicit and aggregate LLM responses for predictions. Experiments demonstrate improvements over baselines, with error analysis on media popularity and region, and a released dataset/code at https://github.com/mbzuai-nlp/llm-media-profiling.

The Geopolitics of AI Safety: A Causal Analysis of Regional LLM Bias

arXiv ·

This study introduces a Probabilistic Graphical Model (PGM) framework utilizing Pearl's do-operator to causally audit LLM safety mechanisms, specifically isolating the effect of injecting cultural demographics into prompts. A large-scale empirical analysis was conducted across seven instruction-tuned models from diverse origins, including the UAE's Falcon3-7B, as well as models from the US, Europe, China, and India, using ToxiGen and BOLD datasets. The findings revealed a disparity between observational and interventional bias, demonstrating that standard fairness metrics can overestimate demographic bias. Western models exhibited higher causal refusal rates for specific demographic groups, while Eastern models showed low overall intervention rates with targeted sensitivities toward regional demographics. Why it matters: This research highlights the geopolitical nuances of LLM safety alignment and the potential for demographic-sensitive over-triggering to restrict benign discourse, which is particularly relevant for diverse regions like the Middle East in developing culturally-aware AI.

Matteo Parsani finishes hand-cycle from east to west coast

KAUST ·

KAUST Professor Matteo Parsani completed a 3,000 km hand-cycling journey across Saudi Arabia, from Dammam to KAUST, over 30 days. The journey, titled “Athar: East to West,” aimed to promote physical activity and awareness for people with disabilities. Parsani visited rehabilitation centers and engaged with people with disabilities, drawing inspiration from Crown Prince's motivational words. Why it matters: This inspiring journey highlights the potential for inclusivity and accessibility within Saudi Arabia, showcasing the nation's hospitality and support for people with disabilities.