AI: A Boon to Humanity or a Threat to Humanity?
 | Author:
Victor Sample
Vic Sample: MT43 News Treasurer |
AI: A Boon to Humanity or a Threat to Humanity?
Victor Sample
Elon Musk says saving for retirement is irrelevant because AI is going to create a world of abundance: ‘It won’t matter’. (Fortune Magazine)
Bill Gates Says 'AI Will Change Society the Most'—Job Disruption Has Already Begun, 'Less Labor' Will Be Needed, And 5-Day Work Week May Disappear. (Benzinga)
Last week, we discussed what AI (Artificial Intelligence) actually does: it analyzes massive amounts of information (data, books, magazines, etc). Then, when prompted, uses that massive amount of data to create text and statistically determine the next word or phrase (or pixels in generating images).
Both Elon Musk and Bill Gates are forecasting a future where AI is doing most of the labor. Elon Musk goes further in stating AI will create a “world of abundance”.
So AI is pretty much perfected and is changing our world! But what are the pitfalls of using AI? There are many!
I recently read an article stating that AI tech company Anthropic has agreed to pay $1.5 BILLION dollars in damages to authors whose work they obtained from pirated sources and incorporated into their AI engine training. Several Montana authors will be receiving payouts from the settlement. Not only was their work illegally used without compensation, training the AI engines received will be incorporated into what the AI engine generates – without crediting the authors.
Plagiarism is a very real concern!
Last November, I asked the Microsoft CoPilot AI agent to generate a family having a Thanksgiving feast. With the famous Norman Rockwell painting in mind (“Freedom from Want", 1943), I asked that the image include both males and females from different age groups. CoPilot generated an incredible image – as a matter of fact, it pretty much was a duplicate of the Rockwell painting, but not as well done.
The Rockwell painting is not in the public domain. When I asked about this Rockwell painting being plagiarized, CoPilot informed me that there were several differences: some of the silverware depicted was different; some of the colors were different; one of the facial expressions was different. According to CoPilot, as long as the image is not an EXACT copy, it doesn’t break any copyright laws. In my opinion, it certainly breaks the intent of the laws! I used the image even with my qualms - mostly because if anyone threatened to sue us, I could reference the CoPilot statements. Microsoft would be a much better legal target!
Another article I read was about a Columbia University professor using AI to determine if student papers were written by AI. He said he became aware of the problem with students generating papers with AI when so many of them were very articulate and were extremely well done. In his experience, not ALL college students were that astute. In querying them about their papers, he found that many of the students didn’t really understand the subject matter nearly as well at their papers indicated. Many of those students didn’t even remember what was actually in their papers.
As we know, AI is generating text based on the statistical probability of the text it is generating – not at all based on actual knowledge of the subject matter. In my next article, we will talk about even bigger issues than plagiarism.
