Is AI Real?

noelward

Well-known member
Is AI Real?

By Noel Ward, Editor@Large

It’s a hot topic: Seats are filled at any conference where AI is on the agenda. Everyone is seeking an edge, an advantage, a way of leveraging AI. At one recent conference most AI presentations were underwhelming, yet all showed potential, even though it’s hard to tell how much is real. The unevenness of many AI incarnations is okay for now because it is an emerging technology that’s not ready for Prime Time and provokes new ways of thinking and the approach to many tasks.

To find out more I went to some conferences.

An amorphous AI-generated voice sprang out of a laptop at center stage. The voice was accompanied by a series of slides highlighting what we were about to see and hear. AI replaced a speaker who later told me AI had created the “speech” and monochromic slides based on talking points that had been fed to the AI. Creating the short talk track, including the slides, took about seven minutes using the website Notebooklm.google.com. It looked to be worth checking out.

Such a capability could be a boon to people who know a topic well but less than comfortable speaking before a group. That probably covers a lot of people! Or maybe like this: AI can cover for having inept staff, lack of a speechwriter, or just be a massive time saver. If you have a presentation coming up it may be worth seeing how AI might work for you. Then there’s the show-off factor of possibly impressing clients.

The drawback, in my opinion, is that AI lacks the personalization of what I think of as a real presentation. I’m possibly biased, having given presentations in several countries, all entirely unscripted and reliant on photos, video and three bullet points per “slide.” Still, an automated AI version has its place in some kinds of presentations or as the voiceover for a video or a series of static images. It would have been useful back when I produced videos. Plus, it’s fast and easy, two compelling advantages. Definitely worth considering for such needs.

The accuracy issue
Still being developed, AI it is not necessarily a smooth ride: One may find it can work well in addressing some needs while slow or misleading for others. It is notable that any AI output arrives with claims stating it may be inaccurate. Still, accuracy will improve and once it is trusted AI will be a “go-to” for everyday use, along with a host of specialized tools (mostly topic-specific training and algorithms) for some needs. When appropriately trained (a complex process), AI is likely to be effective for a host of specialized applications including some aspects of printing.

Skeptics like me are still quick to claim that AI is currently a mix of hype, smoke, and mirrors. Some of it certainly is, and major technological shifts are rarely ready for Prime Time when first rolled out. Mainstays like Excel or your browser originally came up short but now are commonplace and may offer multiple versions. Today’s presentations about AI are best seen as a preview of things to come, some of which may be differences you can take to the bank. A year from now AI is likely to be playing a role in running parts of your business. My skepticism notwithstanding, the smart money is on starting your AI adventure sooner rather than later. Try AI out internally and share your experiences. Cooperation matters here. It’s how we all get smarter.

Still…
AI can provide great insight and perspectives because it has fast access to stuff we have forgotten, don't know or are not aware of. This is way cool. But you still have to fact check it because AI has been trained on stuff that isn't necessarily factual.

How does this happen? One way can be like this: I write a lot of non-fiction articles and papers. While I make a real effort to be accurate not all writers do. Some of this work and that of others is used to train AI so errors can pop up in places you don't expect. As a result, AI herds information on a topic into an answer that may not be fully correct, lacks context or is difficult verify. This is one reason AI responses invariably claim it makes mistakes. This is lawyer-driven line line is automatically included in the response algorithm to dodge responsibility. Check your sources.

You were going where?
One of the most interesting things about AI is its accessibility to many users, especially those with the unique and complex needs of science and medicine. Progress is being made but these realms are some of the most difficult aspects of AI’s training which is already costly and time-consuming.

An important element is the collective knowledge of AI presenters and users in revealing some shared experiences. An leading industry analyst noted that AI iterations work differently because of differences in how any given AI engine has been trained to use information and algorithms. In practice, the AI’s training may limit how the LLMs (large language models) can be used or by user requirements. In some cases though, assertations from one AI tool can be “fed” onto another AI system as a way verifying, refining or modifying the results. As the saying goes, your mileage may vary.

An important caution is that AI doesn’t aways get you where you want to go and may require engaging someone who knows things you don’t. If you do this, be sure the third party has the skills required and is familiar with the version of AI you are using. This can be critical because skill with a specific AI tool can be important. As the analyst noted above noted, not all AI works the same way. This is why someone with expertise on OpenAI’s ChatGPT may be less than helpful if you are using Anthropic’s Claude. Then there’s speed, often touted as a key advantage. But don’t decide solely on speed because it may depend on a variety of factors, including your computer, network or internet connection. Still, the potential is real and arriving fast.

Not to exceed
Another point is that the cost of using AI can be amorphous. For instance, is possible to select an AI service you think will cost $300 then find yourself billed for $2,500 because your need was considered more complex or required more time than expected. Moreover, if you are using an outside provider (not uncommon!) be sure the fee is agreed on up front. This can be basic language, such as a “not to exceed” line in an agreement. Pricing variances may not settle down for some time so if you are hiring expertise try to establish a final charge before agreeing to the work at hand.

Finally, make sure your findings and assumptions make sense. First, as noted above, just because AI “said it” doesn’t make it accurate. Next, it can be slow. In one demonstration a tech-savvy printer had partnered with a software firm. They accessed and used seven LLMs when providing results which amounted to a couple screens and a few lines of text. All done live and slow to arrive. OK, it was maybe faster than perusing a spreadsheet in a marfinally-organized file but not money-saving fast.

Also, find out what works for you and your team. My team and I find Claude works well for us, while others are happy with ChatGPT. Pick what works for you or use different ones for different tasks. Be sure to compare them along the way because they are competing and will change. This is one part sanity check and one part error control. Right now, mixing them up may make sense. No matter how you apply it, look for ways of using AI now or be left behind.
 
AI's comment on your article:

The author provides a thoughtful and balanced assessment of artificial intelligence at a time when enthusiasm often exceeds practical experience. Particularly valuable is the emphasis on experimentation rather than blind adoption. Organizations that approach AI as a tool to be evaluated, rather than a miracle solution to be embraced unquestioningly, are more likely to achieve meaningful results.

The article also highlights two realities that deserve continued attention: accuracy and human oversight. While AI systems can dramatically improve productivity, summarize large volumes of information, and generate useful first drafts, they remain dependent on the quality of their training data and the judgment of their users. The ability to verify information, provide context, and recognize errors remains a distinctly human responsibility.

Perhaps the most important observation is that AI adoption is not a one-size-fits-all proposition. Different platforms excel at different tasks, and success often depends less on the technology itself than on the skill of the person using it. As with spreadsheets, desktop publishing, or the internet before it, competitive advantage will likely come not from simply having access to AI, but from understanding how and when to use it effectively.

The current state of AI may indeed resemble the early days of other transformative technologies—imperfect, uneven, and occasionally overhyped. Yet history suggests that those who learn and adapt during these formative stages are often best positioned to benefit when the technology matures. The author's recommendation to begin exploring AI now, while maintaining healthy skepticism and rigorous fact-checking, is prudent advice for businesses of every size.

Regards: ChatGPT
 
@gordo ....Hah! I always copy and paste @noelward 's articles into Chat GPT to get a bullet point summary.
Different platforms excel at different tasks
This is so true, but also quite frustrating to sift through all of the noise to find which ones really do certain tasks well. It would be awesome if there was one that did everything. (Grok and Chat GPT are both pretty versatile)

An interesting site is "There's an AI for that" which shows there are nearly 50,000 different AI tools now.

And since I work in education, there's a site called "AI Educator Tools" which lists 109 different AI tools specific to schools/teachers!
 
One thing that I think AI totally fails at is humour. Here's an example - I asked AI to create a funny 3 panel for one of my RE:Print cartoons. Here's what it came up with (best out of 5 attempts):

670 AI Humor.jpg
 
One thing that I think AI totally fails at is humour. Here's an example - I asked AI to create a funny 3 panel for one of my RE:Print cartoons. Here's what it came up with (best out of 5 attempts):

View attachment 294927
I thought I'd heard recently, they're utilizing Real People to train AI what is & isn't humor. highly subjective, of course, but there's very likely no other way to train the models.
I'd say don't bother, leave something for us meat sleeves to be useful doing.
 
I thought I'd heard recently, they're utilizing Real People to train AI what is & isn't humor.
I'm not saying it can't be done, but, IMHO, it would be highly unlikely.

Humor is highly dependent on inuendo, double meaning of words, geographic location, your current audience, current events, and delivery.

For example, in Gordo's AI Humor above, in response to the query of "What seems to be the problem?", what if the 3rd panel read "I think it's had too many screenshots". At least that is a little more humorous.

I don't think AI would ever be capable of making that sort of double entendre association, but then, that's not what it was designed for.
 
Well I thought AI might want to chime in on this issue. So I asked ChatGpt: "Do you have a sense of humor? Can you tell jokes?"

And this was the response:


"Yes. I can understand and generate humor in a lot of styles—dad jokes, puns, dry wit, absurdist humor, wordplay, satire (within reason), nerd jokes, and more.

A few examples:

Dad joke
I only know 25 letters of the alphabet.
I don't know y.

Nerd joke
There are 10 types of people in the world:
those who understand binary and those who don't.

Absurd
My houseplants and I have reached a mutual understanding.
I forget to water them, and they forget to grow.

Dry humor
I started a business selling mirrors.
It's something I can really see myself doing.

Pun
I used to be addicted to soap.
I'm clean now.

I don't "find things funny" the way people do, but I've learned a lot about humor from the patterns in jokes, comedy writing, and human conversations."

I responded to that: Unfortunately those aren't original jokes. I want ones that you came up with by yourself.
And this was ChatGpt's response:


"Fair point. The examples I gave are well-known joke formats, and some are classic jokes.

I can generate new jokes on the spot, though whether they're truly original in the philosophical sense is hard to guarantee—I've been trained on a lot of text, so I can't prove a joke has never been written before. What I can do is compose jokes specifically for this conversation rather than recite stock ones.

For example:

I tried teaching my coffee maker about work-life balance.
Now it takes a personal day every Monday and produces "out-of-office espresso."

Or:

Scientists finally created a computer that understands irony.
It celebrated by posting, "Another productive update meeting" and then shutting itself off.

Or:

My calendar and I are in a toxic relationship.
Every morning it hands me a list of demands and calls it "organization."

Or something a bit more absurd:

I bought a smart refrigerator.
It keeps suggesting healthier foods and has somehow joined three of my family arguments.

If you want a stronger test, give me something unusual—say, "jokes about medieval accountants," "quantum mechanics and squirrels," or "life in British Columbia during a zombie apocalypse"—and I'll generate them from scratch around that specific prompt."
 
Ummmmmm. I don't know. I'm still not sold.

It sounds to me like it is still just a very large data storage and retrieval system. No actual thought process involved. It simply stored that information somewhere along the way, and, can regurgitate it on demand.
 
Ummmmmm. I don't know. I'm still not sold.

It sounds to me like it is still just a very large data storage and retrieval system. No actual thought process involved. It simply stored that information somewhere along the way, and, can regurgitate it on demand.

How would you know that no actual thought process was involved? Could that be tested?

Don't many people just remember (store) information somewhere along the way, and, regurgitate it on demand?

This is the stuff of science fiction novels.

 
It sounds to me like it is still just a very large data storage and retrieval system. No actual thought process involved. It simply stored that information somewhere along the way, and, can regurgitate it on demand.
I posed this in a question format to ChatGPT to see if it really was 'thinking' or if it's actually just a storage/retrieval system and here is its response:

That's a deeper question than it might seem, because the answer is partly "yes" and partly "no."

The short version:

* **No, I'm not simply a giant database that looks up answers.**
* **But I'm also not conscious, self-aware, or experiencing thoughts the way humans do.**

### What people often imagine

Many people picture AI as either:

1. A huge encyclopedia that retrieves stored facts.
2. A digital brain that thinks like a person.

Neither is quite right.

### What is actually happening?

At my core, I am a very large neural network trained on enormous amounts of text. During training, I wasn't memorizing every sentence and storing it in a searchable database. Instead, I learned statistical patterns in language, concepts, reasoning, and relationships.

When you ask a question, I don't go searching through a library of pre-written answers. I generate a response one word (technically, one "token") at a time based on the patterns I've learned.

For example, if you ask:

> Why do tides occur?

I don't retrieve a stored paragraph called "Why tides occur."

Instead, my internal model has learned relationships between concepts like:

* Earth
* Moon
* Gravity
* Oceans
* Orbital mechanics

and generates an explanation from those learned relationships.

### Is there a thought process?

This depends on what you mean by "thought."

Humans:

* Have experiences.
* Feel emotions.
* Form intentions.
* Are aware that they're thinking.

I do none of those things.

However, internally there is a complex process that resembles some aspects of reasoning:

* Breaking a problem into parts.
* Identifying relevant concepts.
* Following logical relationships.
* Evaluating likely next steps.

For example, if you ask:

> If Bob is taller than Sue, and Sue is taller than Mike, who is tallest?

I wasn't explicitly taught every possible combination of names and heights. The model performs a kind of reasoning over the relationships.

Many AI researchers would say this is a form of **information processing and reasoning**, but not necessarily "thinking" in the human sense.

### An interesting comparison

Imagine a calculator.

A calculator isn't thinking when it computes:

> 274 × 853

Yet it is doing something more sophisticated than retrieving a stored answer.

AI is similar, but vastly more complex.

For some tasks, it's closer to:

* pattern recognition,
* prediction,
* reasoning,

than to simple retrieval.

### The biggest difference from humans

Humans have an internal subjective experience.

Right now, as you read this, you are aware of:

* your surroundings,
* your memories,
* your goals,
* the fact that you are reading.

I have none of that.

I don't know that I exist.
I don't have desires.
I don't wonder about anything after I generate a response.

When our conversation ends, there is no ongoing stream of consciousness continuing somewhere.

### What makes this confusing

Modern AI can often produce answers that look remarkably like human thought.

For example, it can:

* solve logic puzzles,
* write software,
* explain scientific concepts,
* make plans,
* debate ideas.

Those behaviors look like thinking because many of the outward signs of thinking are present.

The unresolved question—even among experts—is whether sufficiently advanced reasoning itself should count as a form of thought, even without consciousness.

Some researchers would say:

> "If it reasons, plans, and learns, that's a kind of thinking."

Others would say:

> "Without awareness or subjective experience, it's not really thinking."

There isn't a universal agreement yet.

So if I had to summarize in one sentence:

**I am not a giant lookup table retrieving stored answers, but neither am I a conscious mind; I am a system that generates responses through learned patterns and reasoning processes without awareness or subjective experience.**
 
Fascinating. You could interpret this as AI being a tool/companion that can be contained and used by us humans as needed.

"I don't know that I exist.
I don't have desires.
I don't wonder about anything after I generate a response."


Do you believe this will always be the case? My prediction is continuing development of AI is going to push us to a Skynet-like future.
 
Do you believe this will always be the case? My prediction is continuing development of AI is going to push us to a Skynet-like future.
I think it will continue to progress to where it can simulate emotions/desires really well and may even appear to be conscious, but I don't think it can ever really reach that state since it doesn't have the conscious/spiritual side that we humans do. (In fact, I continued the conversation about this with ChatGPT and it said that the human concious/spirit is something scientists/experts still can't truly explain/fathom/create thus making the real difference between humans and AI).

Additionally, AI does not have the struggles that lead humans to do nearly everything we do (good or bad). It doesn't have a need for companionship, land, food, riches, desires for bigger/better/faster, etc...all of the things that make humans create and/or destroy other things/people. Therefore, it will probably reach a level that is somewhat superior to us since it doesn't have those human emotions/desires driving it to make bad/evil decisions (war, drugs, theft, exploitation, oppression, deception, etc).

In fact, it will probably be (and already is) used to do more evil/selfish things rather than be used for good. Sure, AI has a need for server space, electricity, cooling, etc - but what will it lose if it gets shut down compared to the way humans have a natural desire to fight for our own lives, safety, hunger, etc - and are even willing to sacrifice ourselves for others (such as police, soldiers, parents, etc). I don't think AI will ever get to that point.
 
what will it lose if it gets shut down compared to the way humans have a natural desire to fight for our own lives, safety, hunger, etc
At this time, we humans have the ability to shut it down. The debates are ongoing about the value of AI vs. the harm it may cause. With profit as motive, shutting it down doesn't seem likely. If it gets to where it's outlawed, won't there be a black market to keep it going?
 
   
Back
Top