What Is AI And Why Should You Care?

Allen E. Paulson College of Engineering and Computing

AI, or Artificial Intelligence, is the ability for computers to perform tasks that require human intelligence and cognition. Recent advances in hardware and software, coupled with cloud technologies, have facilitated the growth and hype surrounding modern AI. Developments garnering attention are in deep learning, which is largely based on neural networks. Neural networks mimic the human learning process by creating neural pathways in computing. Deep learning has advanced image and speech recognition which has a wide range of applications. One such example is helping the visually impaired to see via smart devices which use image recognition to dictate surroundings.

Examples of AI applications are image recognition, learning and making predictions from historical data, text processing and speech recognition, and even self-driving vehicles. AI touches our daily lives. It aids physicians in making diagnoses by, for example, processing radiological images. AI helps guide investment strategies, and powers recommender systems such as those that recommend movies on Netflix or products on Amazon or filter your spam. Furthermore, AI is being utilized in areas such as facial detection, anti-terrorism, license plate recognition and fraud detection.

While there is a good deal of hype about the scary future of AI and taking over humans — such as the movie “The Matrix” — it’s just hype. AI has been around for more than a half century and our current AI systems are not as intelligent as a canine. Take the recent advances in image recognition using “deep learning.” This is a complex process for computers but very easy for Fido. We won’t venture into the calculations and intelligence to catch a tennis ball. 

One often overlooked area of AI which could have negative impacts relates to cybersecurity, which needs to be considered as AI moves forward. AI is trained on data and if a hacker were to poison the dataset, an AI system could produce malicious results. Like a human operating on incomplete or wrong information, AI is only as good as the data from which it learns. 

Hayden Wimmer, Ph.D., Associate Professor of Information Technology,
Dept. of Information Technology