By William Crooks
Local Journalism Initiative
Editor’s note: This article is a continuation of a series on artificial intelligence (AI). The Record initially spoke with Dr. Stefan Bruda, Bishop’s professor of computer science, who explained that AIs are machine learning programs that mechanically produce the same results as rational agents (human beings).
AI is a tool with potential colossal future benefits, Reena Atanasiadis, dean of Bishop’s University’s Williams School of Business, insists. The Record spoke with her over the phone to get her views on AI’s role in various areas of human endeavour.
AI and medicine
In the medical industry, we know that AI can detect slight aberrations in x-rays that the human eye cannot, Atanasiadis said, and is therefore able to better detect, for instance, breast cancer.
We’ve had “expert systems” for a while, she said, but now they can “dialogue” and produce better “data-driven conclusions.”
Atanasiadis sees the future possibility of the public using AIs to self-diagnose, which could be beneficial or dangerous. Even better, she stated, doctors could use it to diagnose rare diseases that have symptoms in common with more typical diseases; the AI would help eliminate the doctors’ confirmation bias to produce better results and diminish mistakes.
Analogies with other technological advances
“Microsoft [a popular tech company] cut a cheque for $10 million to ChatGPT,” she continued, which is a leading new AI program. This is because Microsoft wishes to add an AI tool into its suite of products, she said.
When spell-checking technology was released, many argued against its use because having correct spelling was part of how students were assessed, Atanasiadis said. We came to realize that what students were saying was more important than their ability to spell if that was perfected by technological innovation.
Making another analogy, she commented on the advent of cash registers that automatically determine what change to give back in place of the cashier making their own calculations. This new technology cut down on human error and improved profitability.
“60 years ago, we only had tellers in banks,” she continued. Now, we have ATMs and online banking, but the tellers are still there.
Imagining a better future
“ChatGPT is going to move the needle,” she said, and the public will use it increasingly in everyday life. We are too quick to worry about “obsolescence” and science fiction scenarios like the Borg (a Star Trek hive-minded group of evil cyborgs).
Atanasiadis looks to philosophers and writers to present us with a better vision of how AI will affect our lives. Existing AI will then have those rosier pictures and human values as a part of their databases, which will affect what they do. “We will defend ourselves [from the negative consequences of AI]… by [making AI] less like machines and more like humans.”
AI and the stock market
It might be thought that AI could affect the stock markets negatively, much like, in recent news, Reddit (a popular online messaging forum) groups led investors to band together and buy unpopular stock, which lost some large hedge funds a lot of money. But she thinks this is unlikely. AI would follow trendlines, she said, and not go against “common sense” to prove a social point. In the stock market, like anywhere else, AI will be a tool that allows us to streamline our decisions for the better. If financial institutions could cut salary costs, they would, but brokers are still necessary.
It is human nature to say “wow, what is [AI] going to do?” she emphasized, but the future is, very likely, not apocalyptic.