Article written by Zayd (Year 9)
My AI Seminar Experience at Imperial College London
I had a rare opportunity to attend a full day talk on AI at the Imperial College London. The day was mind-blowing, like stepping into the future, yet it was there and then!
The visit started on an impressive high note when we found ourselves in a large boardroom at the TRANSLATION & INNOVATION HUB, comfortably seated, getting an engaging introduction to the world of AI, led by a man at the forefront of AI research and development, Professor Oren Kurland.
AI, Artificial Intelligence, is relatively young in the world of science. It started gathering its momentum in the late 90s, when a machine learning algorithm defeated the World Chess Champion and a Grandmaster Garry Kasparov. 30 years later, following rapid technology advancement, AI, inconspicuously, already engulfs us entirely.
If one has ever wondered, how they get YouTube or Netflix recommendations, how the different Apple product ads entice to make a purchase soon after visiting an online Apple store, or how Google seems to know the first morning search would be for the daily commute train schedule, the answer is simple. AI has the capability to use its near unlimited compute resources to analyse our daily digital footprint – the sites we visit, the searches we make, the videos we watch and likes we give, to predict our behaviours through machine learning algorithms and patterns.
We learned about the ‘nearest neighbour’ concept – the fundamental principle of AI and machine learning, that is used to analyse data, determine similarities between different data sets and find the nearest likely match. To answer an earlier question about YouTube and Netflix, the “nearest neighbour” algorithm finds five most similar accounts (nearest neighbours) based on the common shows and movies watched and then recommend other shows from their lists that you may not have yet watched. “People who watched this movie also watched …”, sounds familiar?
It was an eye-opening moment to realise the enormous potential for AI and how its application can change the future:
- AI can revolutionise healthcare by helping to diagnose diseases earlier and more accurately by analysing medical images, genetic information, and patient data.
- AI can personalise educational content to individual student needs, providing customized resources and feedback to enhance learning.
- Self-driving cars, trucks, and drones can reduce traffic, transform delivery services and improve efficiency.
- AI can improve global sustainability through optimizing the use of natural resources, such as water and energy, and improve waste management processes.
- AI can identify and respond to cybersecurity threats faster than traditional methods, protecting sensitive data and systems.
It was also a bit daunting to think that AI, misguided by unethical behaviours, could also be put to misuse, unless additional measures around safety & security, privacy and accountability are put in place to ensure responsible application of AI. We don’t want AI to think that it’s acceptable for a self-driving car to jump the amber traffic light, because it could be an occasional human behaviour, or share medical records of another person based on the fact you both exhibited same cold symptoms last month.
One thing is certain: AI has already wrapped its invisible digital hand around our shoulders, taking us onto exciting and incredible journey. Whether it’s developing AI or figuring out how to use it for good, the future is wide open, and I can’t wait to see where it leads!
