HCAIEP

As social media continues to grow, the prevalence of abusive language on these platforms has emerged as a major safety concern, particularly for young people exposed to such harmful content, motivating our study. We aim to identify and classify instances ...

Over the last few years, using LLM chatbots like ClaudeAI, Co-pilot and ChatGPT for text generation has become a regular habit for many, with over 100 million weekly users flocking to ChatGPT alone. One side effect of such vast usage of these models is ...

Cyberbullying is an intentional aggressive act carried out repeatedly through electronic means against a victim unable to defend themself. This phenomenon combines aggressive behavior, repetitive actions, and the victim's inability to defend themself ...

Artificial Intelligence (AI) is becoming a common topic within the computing K-12 curricula worldwide. While much of the focus of research is on the use of Generative AI in and for education, AI as a core subject area is still gaining popularity, with ...

As Artificial Intelligence (AI) becomes increasingly pervasive in our lives, it is vital to enhance the AI literacy of all citizens. Young adults, for example, must be equipped with adequate understanding of AI to prepare them for the responsible use of ...

Since the conception of Large Language Models (LLMs), their areas of application have increased significantly over time. This is due to their nature of being able to perform natural language processing (NLP) tasks (like question answering, text generation,...

LLMs are being used in an increasing number of AI applications, raising important ethical considerations for which comprehensive guidelines must be framed. LLMs have equal measures of power and risks in bias, privacy breaches, lack of transparency, and ...

This study investigates fairness in facial verification using a Siamese neural network, which is designed to compare two facial images by learning their similarity. The network’s identical Siamese subnetworks process input images to produce feature ...

Hubert Dreyfus questioned the foundational ideas of early artificial intelligence research. Challenging the prevailing orthodoxy, he posited that the failure to manifest advancements in areas like language translation and problem-solving stems from a ...

This study investigates the impact of Privacy-Preserving Machine Learning (PPML) techniques on data anonymisation, with a focus on the risk of re-identification. The key PPML method used is differential privacy, while bias is introduced through Laplace ...

In the current landscape, diversity and inclusion are highly emphasised, this research proposed a methodology to identify and mitigate gender bias in AI recruitment systems. The methodology included identifying biases based on the U.S. 80% Rule, ...

This study examines the privacy implications of using Human Activity Recognition (HAR) features for violent action detection in the fight against bullying. Given the sensitive nature of such applications, these systems must comply with privacy regulations,...

Bullying and cyberbullying are widespread issues that profoundly affect the mental health and overall well-being of young people. To address these matters, we propose an AI-powered tool to group and tag individuals based on their responses to profiling ...

We introduce a case study while teaching human rights, ethics, and risk mitigation methods for engineers. This allows for a hands-on approach to understanding and implementing human rights by teams of students. This exercise has exponentially increased ...

This study targets major problems designers face when designing human-AI systems, such as AI literacy gaps and collaborative design challenges. The current research syntheses present the AI-powered service blueprint, a structured tool originating from ...

This study compares black-box and glass-box models for face verification, specifically using Siamese neural networks and K-Nearest Neighbors (KNN). Using the Olivetti dataset, we showed that a Siamese network with VGG19 transfer learning reached 95.4% ...

In this poster, we discuss popular LLM models- in this case LLM Models like LLAMA 2, Gemma and Mistral 7B- for indirect and direct biases through 3 open-sourced and research-oriented datasets with over 1000 prompts in total on fairness factors like gender ...