Ethnics and Safety

If humans do not understand the problem and solve it themselves, how could machines learn if humans do not even know themselves to teach the machines to solve the problem?

About the term "Artificial Intelligence Hallucinations"

Since my first year in college, a cognitive science research poster of a student in my year at UC Merced during my second poster session has always been in my mind: on how humans follow robots blindly even though the robots are not making the right decisions and leading them into danger.

Quote to Keep In Mind

In October 23rd 2025, I met with my PhD mentor in my second year in MESA to ask for feedback for my predictive maintenance capstone project. The information he typed on the document I shared with him provides one of the best explanations for Artificial Intelligence Hallucinations I have seen.

For instance, autonomous driving with cars such as Waymo offering driver-less cars driven using computer vision and robotics techniques. However on ethnics and human safety, this is where the issues come into play, as for an unavoidable accident between two humans the artificial intelligence needs to make the decision using information gathered through the computer networks and embedded systems: 

Which human's life is worth saving over another?

The main guidance I have been told by high respected Explainable Artificial Intelligence researchers on machine intelligence to train machines through data humans have labeled for the machines to learn from is:

If humans do not understand the problem and solve it themselves, how could artificial intelligence learn if humans do not even know themselves?

Case Studies

Automous Vehicle Safety

For instance, autonomous driving with cars such as Waymo offering driver-less cars driven using computer vision and robotics techniques. However on ethnics and human safety, this is where the issues come into play, as for an unavoidable accident between two humans the artificial intelligence needs to make the decision using information gathered through the computer networks and embedded systems: 

Which human's life is worth saving over another?

The machines can tell through object detection and image classifier if a person is a child or a elderly person, however how would these machines decide whose life is worth saving over another when us humans have ethical conflict on that ourselves? 

A method proposed by Agrecom CEO Stuart Hawling who came into my capstone class's lecture to do an Explainable Artificial Intelligence alongside Ethnics in Artificial Intelligence talk who using social credit score determined by the Big Data received through Internet of Things (IoT) smart vehicles and smart traffic management.

with Blue Tooth Sensors in traffic intersections with traffic lights collecting data and the Geographic Information through GPS on the car. (last Monday the monday before 11/17 2025)