Artificial Intelligence and HAL
Artificial Intelligence---“I'll get around to using it as a subject of an essay…eventually.” Eventually arrived this week. If you expect to learn much from this essay, turn back now.
If I'm interested in a subject which might be suitable as an essay for Old Art Guy about which I don't have sufficient knowledge, I turn to the internet, print out a number of related articles and spend some time reading the articles and making notes and trying to wrap my old brain around what I've researched.
Of course, I followed the same path with the subject of “artificial intelligence.”
Truthfully I didn't find much information useful to me or the potential reader.
AI in general is the ability of a computer or a machine to think and learn. It's a program that mimics human cognition. The ideal AI program is one that can learn, solve problems, and can optimize the chances of success with some goal or objective.
The ultimate goal of AI developers is producing one that has the flexibility to interpret its environment and take actions to maximize success. Ideally it will be able to learn, solve problems, and think logically. The ultimate goal of the developers is to develop programs of general artificial intelligence that can solve many problems instead of being limited to specific tasks…you know, like actual people only faster and better.
And this is where HAL enters the picture. HAL, a very intelligent, rogue computer which decided to take control of the space mission the humans were on in the motion picture 2001: A Space Odyssey. In the intervening years, our talking computers have sometimes surprised the developers with unexpected behaviors but so far have not become as malicious as HAL. Elon Musk has recently sounded the alarm that developers of AI should slow their roll on the development of the ideal AI program because it may lead to unexpected and tragic results. I heard an AI developer who has witnessed unexpected behavior by his company's AI programs admit that he didn't understand why or how this happened. When questioned about the wisdom of releasing such a program on the world, he responded that he didn't understand the human mind either. Not reassuring.
Of course the AI industry will not slow down or be more cautious with its programs because there are so many developers grasping for the golden ring...even Elon Musk has entered the field of play.
Who are the biggest players in the field of AI? Well, there's a lot, a bunch, a beaucoup of players most of which you probably have never heard. However, I'm sure you're familiar with Google and Microsoft, huge players. In a FOX interview, Elon Musk stated that currently some AI programs are capable of producing extremely persuasive documents which can alter their arguments to be even more persuasive via feedback from the public. He seems to expect that Democrats will use such programs to mislead the American electorate. (Clearly he's a racist.)
Some of the latest AI programs on the edge of new developments are the following: ChatGPT can understand natural language and generate a human-like response. GTP-4, is a multimodal AI that can process not only text but also audio and images. Watson is an AI system which performs well in a number of industries but does a particularly good job in healthcare applications: it can predict and detect various kinds of diseases like cancers, cardiovascular disease, heart disease, etc. with great accuracy and can also recommend medication. We cannot leave out Sophia, a humanoid robot designed to mimic human behavior and appearance…and does it well: in 2017 Saudi Arabia granted Sophia citizenship---gimmicky but good PR.
For those of you who have little or no interest in the future of AI, let's talk about the present day capabilities of AI.
It is very common today for folks to employ virtual assistants. Different products perform different tasks: answer questions through voice commands, setting reminders, playing music, and controlling smart home devices through voice commands. Some of these devices you've probably used or at least have heard of: Siri, Alexa, and Google Assistant.
The business world is gobbling up AI primarily because it can perform tasks that free up its employees to perform less mundane tasks and to some extent predict future business outcomes.
While doing my research, I didn't find much about self-driving cars which is definitely an AI development. It's my understanding that auto companies are improving this product. Tesla is a leader in this area.
My first reaction to self-driving vehicles was, “How friggin' lazy have we become!” Over time I've developed some appreciation for the product…mainly commercial applications for delivery of goods which brings to mind drones which are also dependent on AI for the same purpose.
By the way, I was a little shocked to learn that some of these AI programs if given a beginning sentence of a fictional story, will in a few moments generate a complete story based on the opening sentence, and the story will appear to have been written by human writers…without the grammatical errors or misspellings. Just what I need, more competition.
The AI industry is moving so quickly, it's difficult for the analysts of what's happening in the industry to keep up. Most of my research was less than satisfying. I suspected that many of the writers of the information about AI were trying but truly didn't understand what was going on in the AI world. Bless their hearts. Well, at least this realization made me feel a little better about my own lack of knowledge about AI.
We're just in the earliest stages of artificial intelligence. I say, “Go for it!”…but keep an eye out for HAL.
enough