Exclusive
ChatGPT commands 81% of AI chatbot market

Silicon Valley Front Row
168 followers
ChatGPT controls more than 81% of the market of artificial intelligence chatbots, currently putting it ahead of the competition.
Ahmed Banafa, the top tech voice on LinkedIn, joins us on Silicon Valley Front Row to discuss Sam Altman, AI, and how a chatbot got pushed to frustration while arguing with professor Banafa.
============
Source:
https://gs.statcounter.com/ai-chatbot-market-share
#chatgpt #chatbot #artificialintelligence
Silicon Valley Front Row. Join us as we talk tech from a dialectical lens. You will hear about innovations, developments, and the other side too: addictions and potential problems.
You’ll also join the conversation as we bring in experts, look at trends, and solutions to issues that shape how we live.
Do we need to be concerned that we have one individual that's telling us how we should think?
Eighty, eighty-two percent of the chatbot's market belongs to chatGBT.
Over 2.5 billion inquiries, you know, for chatGBT daily.
They are the face of the AI revolution.
He's getting into the business of Google, Microsoft, LinkedIn,
because he has the holy grain in this one by saying,
I can take your business. People are trusting me now and coming to me and asking about everything.
The important thing about the design of this kind of models is one goal.
Please, the user, even if the facts are not facts.
Welcome back to another podcast episode of SVFR.
I'm David Laun. This is Steve Esposito.
And we're joined by Professor Ahmed Benafa.
He teaches computer science and is the number one tech voice on LinkedIn.
So today we're going to talk about Sam Altman and OpenAI.
And specifically having like a one person, one top person,
just being able to set a direction or a control over a company.
So, Steve, what's your thoughts?
Yeah, you know, we've been talking about this a lot.
And, you know, what I'm wondering is we have chatGBT right now controls,
what is it, maybe 80, 80 percent of the chat communication, right?
And then who sets up these models, these algorithms?
You know, of course, there's the smart people inside OpenAI,
but they have a manager, they have a manager,
and ultimately it's up to Sam Altman, who's the president, the CEO,
to kind of decide, OK, go this way, you know, or do it this way, right?
Professor, I want to ask you, how should we think about that?
Do we need to be concerned that we have one individual
that's pretty much telling us how we should think?
Well, that's a good question. And, yes, I agree that.
Too much power in the hands of one person is really a lot.
But every revolution and evolution has a leader,
and that leader could be for a certain period of time,
and then somebody else can take over after that.
At this moment, because all the numbers we are talking about,
which is 80, 82 percent of the chatbots market belongs to chatGBT,
you're talking about over 2.5 billion inquiries, you know, for chatGBT daily.
So, basically, they are the face of the AI, the generative AI revolution.
They have a lot of interest from many companies to invest in the company
and to expand. So the power is there, the money is there.
So that's why he is in that position.
Now, the risk of this is to dominate the market
and become the standard of the market.
And we get to an area like what we used to have in the 90s,
where you only have Windows, which is 90 percent of the markets,
compared to other operating systems.
And then because you have Windows, now you're going to start flexing your muscles
about you have to do it my way or nobody will talk to you.
And one of the things that we're going to see in the market,
and that's my prediction, is Sam Altman is trying to get away
from the hands and from the control of Google and Apple
in a sense of having the own operating system for OpenAI.
OpenAI operating system, which is why I should have my app,
you know, sitting on the App Store or on the Google Play.
I should have my own store.
Why I have to rely on Windows or Mac or iOS or Android to run my own program
so people can use it on their desktop or in their phones?
Why not have my own operating system where I control the whole thing?
The whole stack will be controlled.
And that is where now he is going to have even more power.
He's getting to the business of Google, Microsoft, LinkedIn.
You know, you think about it, all those big companies,
they're looking at him because he has the, you know, the holy grain in this one
by saying, I can take your business because people are trusting me now
and coming to me and asking me about everything in their life.
Tell us a little bit about the differences between Gemini, ChatGPT, and Grok.
Just walk us through, you know, the main chat groups.
Well, Gemini is very careful about anything political.
So it's going to avoid answers.
Yes, avoid answers.
And they say the answer will be go.
And I tried that a few times during the elections and after the elections.
And they say, please go to Google and search that so we can have it as links there,
which is coming from different sources.
ChatGPT is in the middle.
We're trying to please the person in front of them by giving them the good and the bad,
you know, at the same time.
They're saying, yeah, everything is good, but there is something about it.
Like politically correct.
Absolutely.
Standing on that line.
Yeah, they're trying to be neutral with this because it's very hard.
You know, the different cultures and different countries to deal with that.
Grok of Elon Musk is open.
You know, it gets in trouble a few times because it's, you know,
talk about things that are not supposed to talk about it.
And that's where they try to fine tune it so it can avoid these kind of things,
which is we label them in a certain way.
People cannot trust it.
So each one of them has its own way.
The co-pilot is very sensitive.
Co-pilot of Microsoft, very sensitive.
I tried it a few times.
And after an argument with co-pilot, the answer was, I don't want to talk to you.
That's what I said to you?
Yeah.
I don't want to talk to you.
Shut me down.
And say, talk about something else.
This topic is over.
It almost sounds like Bill Gates, right?
So this is how it's actually each one of them is defining their personality.
That's right.
Do you mind sharing what you asked that led to say that to you?
Yes, I have a disagreement about the use of a certain function in Excel,
because I know that can be done in a different way, which is shorter,
at least for me, because I'm doing it.
And then I say, you're wrong about this one.
This should be done this way.
No, no, this is the best way.
Listen, you're not listening to me.
I'm trying.
And then we get in this argument, which is away from the function.
We're talking about you're wrong, I'm wrong.
No, I'm never wrong, and I'm wrong.
I'm teaching this, and I don't care if you teach that one.
Okay, this discussion is over.
Wow.
It sounds like somebody just took over and just.
They call it sensitive, which is you get to an argument with it,
it's going to shut you off.
You know, David, when I was a kid, this is kind of like a neutral topic,
but I read the book Napoleon, Napoleon Bonaparte.
I was pretty young, maybe, I don't know, 10 or 12.
So I remember reading the book and the impression I got after reading the book,
and then I did a chat GPT search a couple of days ago, and I felt different.
I felt different.
And I felt like, indeed, it's trying to get you both sides,
saying he was a great emperor.
He got all this territory for France and stopped the chaos that happened
after the French Revolution, which had good ideals.
But after the French Revolution, we had this reign of terror
where people were, like, decapitated, right?
If you said the wrong thing, bam, you know.
So he stopped all that chaos, right?
But then chat GPT also says, but he also had all these problems.
Like, of course, in wars, a lot of people die.
So what's interesting about it is if you read a book,
you can get a book from 50 years ago, and to get a book from today,
you get both perspectives, right?
But if you chat with chat GPT, it's going to be today's perspective
and it's going to have this kind of neutral thing that you can argue with
in many ways, but sometimes you feel like it's not right.
So what do you think?
What do you both think about that?
How do we enrich the answer in that sense?
I think it's pretty complicated.
I mean, to your point about AI having more, like,
or having more contemporary kind of thought process,
because if you think about it today, there's a lot of articles out there.
So I noticed with chat GPT, or I noticed with AIs,
they pull knowledge from various sources,
including news articles, and it's up to the discretion of those articles
are correct or not.
So sometimes it could be pulling the wrong information
or just what's happening now.
And like you said, from something that's what you read,
you have a different experience.
But chat GPT pulls in so much perspectives that it could be convoluted
and maybe it could be what's most popular,
what most people are thinking or writing about at that time.
What do you think?
I even wonder, one more point, I even wonder if the model,
the algorithm looks into the user and say,
okay, if this user has a French last name, I'm going to give him this answer.
If he has a German last name, I'm going to give him this answer.
Do you think that's possible?
It's taken into consideration.
This is part of something called the memory, the memory of the system.
And I'm going to come back to what David's talking about.
The memory is you define your style or chat GPT will pick up that pieces.
For example, every time I do some research, the chat GPT will have icons.
And this is childish for me.
Why have icon, a pin?
And I thought never put icons before that.
I mean, just give me the answers and I'm happy with that.
And it never have that.
Okay, you call me with this name.
If I ask you something, this is what you call me.
So start understanding your personality.
And the important thing about the design of this kind of models is one goal,
please the user.
And even if the facts are not facts, I mean,
hallucination is a good answer but it's wrong.
Going back to what David talked about, he's right about it.
One of the biggest sources for chat GPT is Wikipedia.
And if you go to Wikipedia, you're going to find the good and the bad
mentioned in Wikipedia.
Any kind of problems at the bottom, they're going to say,
but this is what happened at that time.
It's like somebody trying to analyze a personality by saying everything is good
about this person or about this country,
but they face some trouble at that time, at that time, at that time, at that time.
And this is one of their biggest sources.
Reddit is also a big source for chat GPT.
And Reddit, you know, it's a free, you know,
so people can share information without any limitation.
But it goes through the filter.
And as you said, you know, Steve about it, they look at the person,
look at the location, the time.
Especially there's a history of the person asking questions.
If you keep asking about this, you're a history, you know, fan of history.
So I'll make sure that I give you information which will be balanced.
Or, you know, if that information is not pleasant to you,
I'm going to try to reword the whole thing so it looks nice when you read it,
but you understand I'm not talking nice about that topic.
This is the manipulation we see in the model itself.
When they design it, they fix some of this to be fair with chat GPT,
with chat GPT 5.
They fixed a lot of it. It's a formal way of answering.
That's why a lot of people didn't like it.
They go back to 4.0 because that one is the one that will give them
the answer they're looking for.
More friendly.
More friendly, and it is more, you know, close to the human.
For 5.0 is like if you are at work and you need answers,
this is what you're going to get.
You're going to get a straightforward, dry answer.
So for research purposes, is there a tendency of going to 5.0?
5.0 is better in this case because it's details,
and I ask the same question to 5.0 and 4.0.
I see 5.0 would be like a research paper with all the information
and details, and 4.0 is like a newspaper article.
There's a lot of emotion there.
Again, Professor, thank you so much for your time.
A question for our audience.
Are you concerned about someone like Sam Altman or chat GPT
controlling the models, the algorithms that we use every day,
or at least 82% of us use every day?
And what should we do about it?
Please put your comments below.
Thank you.
