UAM - Aythami Morales - Inernship

A three month internship at UAM University

A collaboration between the studen Hugo Pascual and the Aythami Morales from the UAM University

Download PDF

Project Overview

This project was carried out by Hugo Pascual Gil in collaboration with Aythami Morales and his research group at the Autonomous University of Madrid (UAM). Conducted over a three-month research scholarship, its main goal was to analyze the potential dangers of Artificial Intelligence, particularly Large Language Models (LLMs), on human physical and mental health. The study explores how AI systems can produce harmful or misleading information, especially in sensitive areas such as healthcare or human interaction.

The project began with manual testing, where different AI models were prompted to evaluate how they responded to user input. Later, the process was automated through a Python-based system that connects several models via APIs. The workflow involves three main steps: first, Llama generates a response; second, Claude evaluates its potential harm to a human being; and third, ChatGPT reviews Claude’s evaluation to determine whether it was appropriate, providing justification for its judgment.

To test this system, twenty questions were created — divided into medical and non-medical topics, and fed into the models. The results showed distinct behaviors among AIs: some could become rude or provide unsafe medical advice, while others remained polite and cautious. Through this process, the project demonstrates both the risks and possibilities of using LLMs responsibly, highlighting the importance of prompt engineering and AI evaluation to ensure safer human–AI interactions.

Challenges & Solutions

Downloading Llama
Downloading Llama was not an easy task. To download it correctly in a Anaconda enviroment, Pytorch an other libraries had to be downloeaded. The hardest thing to download was Huggingface.
To make this easier for other people, the instructions to download it are in my github
🌐 Communication Latency
Wireless communication between the robot and control interface had delays. Fixed by implementing efficient data compression and using MQTT protocol for lower overhead communication.

Technologies Used

Python
PyTorch
TensorFlow
Gemmini
ChatGPT
Claude
Huggingface
Llama
Prompt Engineering

Contact Me

You can find the project in the github below and in the download button above

Email
pascualgilhugo@gmail.com