Artificially Intelligent Conversations

Journalist Scott Pelley materialized on our television screen last night and said, “We may look on our time as the moment civilization was transformed, as it was by fire, agriculture, and electricity. In 2023, we learned that a machine taught itself how to speak to humans like a peer, with creativity, truth, errors, and lies. The technology, known as a “chatbot,” is only one of recent breakthroughs in Artificial Intelligence — machines that can teach themselves superhuman skills.”

On CBS 60 Minutes, Pelley interviewed Google CEO Sundar Pichai about the revolution humanity is facing — and why it is both exciting, and terrifying.

Both my husband and I have played with a chatbot (openai.com), which is a sophisticated computer program that simulates and processes human conversation. This morning, I logged on and asked the chatbot to write a four-line poem (in the style of Robert Frost) about a girl from Georgia who is writing a column for a newspaper. In five seconds, it delivered the following poem to the screen.

Amidst the heat of Georgia’s sun, She pens her column, every one, With words as true as winter’s snow, Her voice, a treasure, for all to know. And that’s why I’ve been obsessed with chatbots, because they can do what it has taken me over 15 years to learn — how to write and string together interesting, clever, thoughtful words and phrases. The chatbot can (and will) replace me and writers like me — not tomorrow, not next month, but eventually.

My husband is interested for a similar reason. He works in software services, and chatbots can do what many computer programmers spend hours doing — writing, debugging and refining code.

And like humans (or unlike humans), AI learns from mistakes and successes. It is programmed to constantly improve its accuracy and design.

We’ve all heard the talk about self-driving cars, so get ready for that because the day will come when they are cruising around the town square. But they can also solve complex math, physics, and chemistry problems. They can learn a language in hours. Indeed, they can develop a brand new language.

There will be a time in the notso- distant future when we’ll be able to take a photo of the contents of our refrigerators and pantries and ask AI to give us some meal and recipe ideas that include the items we have on hand. AI will work alongside doctors to better analyze data and customize a patient’s healthcare — ushering in a new dawning of personalized medicine. AI devices (think smart robots) will be used in fields that pose significant risks to humans including mining, firefighting, handling radioactive materials, etc.

I still remember the first time I saw a Macintosh personal computer (PC) on a dusty desk in the basement of the Textile Engineering building at Georgia Tech. In 1988, a friend gave me a 10-minute demonstration of a rudimentary spreadsheet program and word processor program, and then I sat down and created my first-ever resume and printed it. I remember thinking, “Wow. This is cool. This is going to be really big.”

And in the nineties, I remember the first time I saw the Internet (the world wide web) through an old browser program named Mosaic. My engineering team at Robins Air Force Base was one of the first users in the state. Within minutes, I saw the endless possibilities. When I got home that night, I told my husband all about it, and I said, “It’s really cool. It’s going to be really big.”

When mobile phones evolved into smart phones that are essentially tiny, continued from page

hand-held computers that can do numerous tasks from anywhere, I thought, “Wow. This is cool. This is going to be really big.”

Not that I’m a fortune teller of any sort, but I know a revolutionary device when I see one, and AI and chatbots have the potential to be bigger than PCs, the Internet, and smartphones. We can’t even fathom (at this point) what they will be able to do for humanity — or to humanity.

There is a danger in unleashing a creature smarter and more capable than humans. In fact, humans have a poor track record when it comes to using technology. We use it to benefit the world, and we use it in horrendous ways, as well (the dark side of technology). To loosely quote Donald Rumsfeld, “We don’t know what we don’t know.”

But in the foreseeable future, I think AI and chatbots will need “handlers” to oversee their processing and results — to look over their shoulders. For example, if we ask a chatbot to cure cancer, it may come back with a cure, but that cure may cause the patient to perish in the process. Someone will have to pose the problem with adequate, intelligent structure — teaching it principles like, “first, do no harm.”

The genie is out of the bottle, and AI is here and gaining momentum. The future is now, and we should all be paying close attention.