AI & You, one more time…

Yes, yes, I know it’s the holiday season, but this can’t wait. Just one more attempt to inform and update. The last one of the year, I promise! After this one, nothing but holiday spirit and merriment! No, really!

I just learned that Sports Illustrated used AI to write an article… and got caught. They denied it, of course. Funny, it has occurred to me that if I keep writing my blog until I’m, say, 93, which is how old Clint Eastwood is now, I would be one of the only writers left whom the readers could trust… trust that they were reading something that a real person wrote. Let’s see, when I’m 93 that would be in 2036.

Can you imagine how famous I might become? Yes, famous, simply for continuing to be a live person who had an opinion, and resisted using AI for ease and personal gain. Think of it! In 2036 my son, Dillon, will be 50 years old, and will finally be able to see his old man be the hero I always told him I was.

It boggles the mind. In only a year or two, we will no longer know whether we’re reading something written by a real person, or a totally human-sounding article derived from AI. Hell, some of us don’t really know now.

For example, I decided to ask chatGPT a question, so you could see how far it’s come already. My question to chatGPT – “How long until we are totally unable to distinguish between a human writer and AI?”

chatGPT’s response –

ChatGPT
“Predicting the exact timeline for when artificial intelligence will achieve complete parity with human writers in terms of written content creation and thereby become indistinguishable is challenging. As of my last update in January 2022, AI-generated content has advanced significantly but still lacks the nuanced understanding, creativity, and context awareness that humans possess.

The rate of progress in AI is rapid, driven by advancements in machine learning models, natural language processing (NLP), and deep learning algorithms. However, achieving complete parity with human writing involves not only linguistic accuracy but also the ability to generate content that encompasses emotional intelligence, creativity, empathy, and cultural context—all of which are complex for AI systems to replicate convincingly.

Currently, AI-generated content often excels in specific areas, such as generating news articles, summarizing information, or creating simple stories or essays based on patterns learned from extensive data. However, crafting content that genuinely captures the richness of human experiences, emotions, and creativity remains a significant challenge for AI.

While it’s difficult to provide an exact timeline, some experts speculate that achieving complete indistinguishability between AI and human writing could take several more years or even decades. It depends on breakthroughs in AI technology, as well as our evolving understanding of human cognition, language, and creativity.

Ethical considerations and regulations regarding AI-generated content also play a crucial role in its development and deployment, and these factors may affect the pace of progress toward achieving complete parity.

As of now, AI-generated content often lacks the authenticity, originality, and depth of human-generated content. While AI continues to improve and surprise us with its capabilities, predicting the exact point of indistinguishability remains uncertain.”

******************************

Pretty human-sounding, no? AI is truly powerful, more powerful than we know, even at this early stage. I read an AI blog from England that posts daily updates on new start-ups and new barriers AI is breaking down. AI can write code, design and develop web sites, even duplicate itself and allow the user to put their own spin on its abilities. That makes it a truly dangerous tool in the wrong hands, and we all know there are a ton of wrong hands out there.

A few of the major new companies are attempting to build in some modest guard rails on the new AI releases and chatbots. So far as I can tell, however, no one has come up with a program that can and will recognize any AI-created writing. This is, of course, a huge problem for high school and colleges teachers and professors who are charged with knowing whether the papers turned in for their assignments are AI generated or not. And it’s easy to imagine anyone willing to lie about things will have a field day with AI.

The above AI explanation of itself says, at one point, “… achieving complete parity with human writing involves not only linguistic accuracy but also the ability to generate content that encompasses emotional intelligence, creativity, empathy, and cultural context—all of which are complex for AI systems to replicate convincingly.”

Yeah, well… it also states that it’s last update was in 2022, and the updates, improvements and additions to it in the last year are off the charts. Microsoft, Apple, Google and host of others are coming out with their own versions of AI, all designed for and directed to their perceived high-number consumers. And those abilities it claimed to lack… emotional intelligence, creativity, empathy, and cultural context… are already blurred by the incredible speed and depth of continued development.

How many of us have the time (and actually take it…) to decipher the emotional intelligence or empathy of our information outlets? And even if we did, AI can now virtually duplicate a writing style, which, as far as I can tell, leaves us little choice except to make a law insisting on adding an AI tag to all writing,  art forms, along with human voice and visual duplication. With that should come an almost unreasonably high fine for failure to adhere to the tag.

Perhaps some of this sounds a little silly to you, but it wouldn’t if you are aware of the speed in which this new technology is developing, and how many savvy techies there are who are already developing AI strategies for their own personal, and perhaps greedy, agenda. And I can tell you this for sure, it’s going to happen fast. Some of the smartest (and some evil) people in the world are all over this, and its effects are going to hit us fairly soon.

The European Union (EU) has just come out with some “sweeping” AI guidelines. They want to stop all facial recognition, voice duplicating, certain kinds of job automation, mining the writings of others, and more. As all this falls into better focus, i will update you, if necessary. I care about al this as I understand how powerful the media is, and how a rampant AI might affect it.

For a wake-up call on all this, for a reality check and for checking on me for some reasonable proof of content, Google Ben’s Bites for “top product launches, research and news from the past 24 hours in a 5-min daily digest.”

I, for one, am all about watching and waiting, knowing that my left-handed approach to music and writing will continue to stick out from any AI attempt to thwart me and my still-human need to communicate to… almost anyone.
Definitely NOT AI-generated, I am
Steve Hulse

Leave a reply, always happy to hear from you

This site uses Akismet to reduce spam. Learn how your comment data is processed.

© 2011 - 2020 Steve Hulse, All Rights Reserved