Levels of AI, Trust, and Human Collaboration

Welcome to wake up with AI, the podcast where human powered meets AI assisted. Join your hosts, Chris Carillon, Niko Lofakas, and George b Thomas as we dive deep into the world of artificial intelligence. From the latest AI news to cutting edge tools and skill sets, we are here to help business owners, marketers, and everyday individuals unlock their full potential with the power of AI. Let's get started.

Good morning. Happy Tuesday, November 19, 2024. It is time to wake up with AI. George Nikko, how are you guys doing today?

Doing good. Doing good. I was just I don't know if that robot voice made it onto the show or not, but I was just, reliving my youth in the the war games movie back in the day. Do you want to play a game? It's interesting how smart things are going.

I don't know how smart I'm going, but things are going smartly, so I'm doing good.

Good to hear. How about you, Nico?

Likewise. Doing well. If you guys got a chance to I I am sorry I missed yesterday, but my little girl hadn't been to the groomer in weeks and it's been a nightmare trying to find a groomer in this area for small dogs. So thankfully, we found a good one yesterday, so hopefully everything's cool there. But, yeah, a lot of news over the weekend.

A lot of news yesterday. A lot of very cool interviews, really. So, like, news to me comes in a lot of different ways. It could be actual breaking news, development, something that happened, but it could also just be that somebody gave an interview. Because it's really important to me when these guys give interviews because they don't very often, And they're, again, at the forefront, so they're willing to talk about this stuff very openly.

So a lot in that regard.

Yeah. I was watching a couple couple of them last night. What came to mind for me is that wherever you are in your daily life, your journey, whether it be, like, you know, in your career or, you know, what's important to you at that moment, maybe season might be the right word for some people, but also, like, where you are in the world. And as I was watching, I was listening to the interview with a Microsoft AI person, and he mentioned, like, Copilot, like, going ham in India, basically. And just understanding how different parts of the world are adopting these tools and the more accessible they become, that's gonna happen.

And how does that stuff mix? Just super interesting to think about.

Yeah. I mean, well, you have to you fundamentally have to think about humans. Right? Like, India has been code central for many, many years. So would AI usage be up there based on that coders are doing extreme coding?

Like, that's the thing. I'm really starting to try to pay attention to patterns of historical normal world, like, duh, and then watch how AI is, like, happening in areas so you can be like, duh. Those are those types of humans in those types of areas. But, yeah, I love the interviews conversation, Nico. A couple that you sent over this morning, some clips, then also the longer one of Moe.

Some really interesting things to think about, some some deep insight. I love sitting back and watching and listening to those and kind of allowing myself to dream of a possible future. Sometimes fear a possible future, but I always try to stay in the positive. But I just I love the ability to have time to at least focus or think about these things that you're throwing over across in our Slack channel.

Yeah. And I thought 2 interviews in particular out of all the ones that I've been watching lately were the most profound and for a number of reasons. Some of which personal, others professional, But what George is talking about is a interview that Mogadot did very recently with Sean Kim. And in this interview, he talked about everything across the spectrum. And Mogadot is not afraid to talk about anything in relation to AI.

He doesn't it doesn't matter what the subject is. I really love the fact that he's willing to go into the nitty gritty. He doesn't shy away from questions about, you know, very obvious questions we would have about intelligence and things of that nature. So I did want to just play some clips because we have been talking about this stuff in the last few weeks. To me, it's really interesting to, again, just see how well aligned the show is with some of the experts at the forefront.

So just this is, just one clip of Mogadot talking about the intelligence of AI and, like, what the capability is like.

So so the the the intelligence of a machine will double every 5.7 months. Now that is quite significant. If you just take an IQ, not number of 150, you expect that within 6 months of work, sometimes 8, sometimes 4, but, you know, let's say, by by year end, you're you're gonna get, 300, and then by year end, you know, by 6 months after you get 600, then 1200, then 24100. It's not just adding a 150. It's doubling.

Okay?

The other aspect of it was, like, what was you know, what does that look like? Right? So what does it look like when you have that level of intelligence?

I mean, if if I gave you if I gave you 400 IQ points

more, Sean,

what what would you do? Like, 400 is very significant. Like, you can invent iPhones out of thin air with with, with nanophysics if you have 4 I 400 IQ points more.

So I'd love that part, by the way, because I started to think about, like, iPhones out of thin air. Right? And and you start to think about, like, the incredible power that we as humans could start to have and by the way, when I got to that part, I was like, it's so interesting that for years years years, we've talked about how we only use 10% of our brain. And for years years years, we've been okay with only using 10% of our brain. But yet now we sit here and we go, but if we could have this additional piece and I'm and the nerd in me goes, like, I'd be a freaking hair bender.

Like, I'd, like, I'd be doing some crazy stuff with, like, fire and water and what but, like, out of the ether. I hope everybody's paying attention to, like, that's not a small statement.

You know, the average person is, let's just say, like, around a 100. So, like, where is everything now? Everything now is somewhere around 150, right, which is Einstein level IQ.

Not me, in other words. Just saying.

But that's, you know, that's also using an improper measure too. Right? Because it's saying again these, like, straight lanes of knowledge. Right? It is really good at doing these straight lanes.

What's interesting, you know, we've been talking about AGI lately. We've been talking about, like, whether we got there or, like, how long it's gonna be. I was looking at a, interview with Demas Hassabis from Google DeepMind, and it was him and Dworkish Patel. And they were talking about scaling laws and superhuman AI and, like, timeline for AGI. And Demas says within a decade, that was 8 months ago.

That interview was 8 months ago that Demas said within a decade, and now we are debating it really. I I think the debate is already over. It's well, not yet. It's it's getting there. Right?

We're at 50% on the ARC test with the best human, I think, getting I I misspoke yesterday. The best human, I think, was 98% on the test. So the average human was 80%. So this thing in in a very short amount of time and, like, just in months has gone from 30 to 50%. I'm sure the next jump will be, like, 50 to either 75 or 80 around that that mark.

And we're talking about within easily within a year, if it hasn't already. If what we are seeing at end user level is that, right, then what is going on inside and under the, you know, underneath the roof of these companies is definitely beating the Arc test. Like, it's def it's not I wouldn't say beat, but I mean, it's definitely hitting close. Let's the guy who came up with the test stepped away from Google to start his own thing. Sam Altman said in an interview a week ago, we know what the road we know what the road map is to get to AGI.

And this is a person who said that it was probably gonna happen, who said AGI was gonna happen in 1,000 of days, now says artificial super intelligence is within 1,000 of days. Mocha dot changed his timeline from 2040 to 2037, which I think is still overshooting it. So it's it like the again, the rapid expansion of this stuff. You have to and so it's like, okay, well, they're getting smarter, all that kind of thing. They're getting more powerful, but doesn't necessarily seem like they're getting, you know, generally smarter.

That's where the curveball comes in. MIT just released a paper that talks about post test time training and how that is helping to develop. I gotta say this was this of all the things that I thought that it that was gonna take a long time and technically has taken a long time in digital intelligence world. Abstract thinking. I thought that was gonna take a really long time.

I thought that was something that we were just super unique at. It's just, like, taking a hard left whenever. Right? Nah. Nah.

Got got broken down to math. One of the things that I wanted to head on that, unfortunately, was something that, like, that, you know, we've talked about a a few different times and something that I know that George's head on and kind of speaks to this point was when it came to, like, processing, like human type processing. And so one of the things that is the most abstract for humans to to deal with is feelings and emotions, because they don't really make logical sense, right, according to Spock, but I I feel the same way. So this is a clip from the same interview with Mogadot where he is talking about AI with emotions.

You really have to start humbling humanity a little bit. As a matter of fact, I I not only say that AI is capable of feeling all human emotions. I'm I'm doubting love still, but all of the other rational emotions, if you want. I'm even saying that AI will have a wider spectrum of emotions, simply because they will be a they will have the intellectual, bandwidth to ponder concepts we're unable to ponder. Like, just like we can feel emotions like optimism, pessimism, and hope, okay?

Because we have a concept of the future and, a worm or a butterfly doesn't have those feelings and emotions, And, you know, a a cat probably doesn't because it cannot under the future. And you really have to start

So I thought that was, pretty relevant just in terms of, again, talking about, like, where the range of capability is going as well. You know, like, just what does all of this mean in terms of, like, is just getting smarter. It can, you know, write better essays and, you know, it can help out with medical, which by the way, I take a little bit of joy. There is a little bit of joy in these stories, George. Just a tiny bit.

So turns out just a little bit of a sidetrack that with the latest medical GPT, you know, they're they're starting to pair it up with doctors in order to do diagnosis. And with these diagnosis tests that they're doing, doctors score on average somewhere around, you know, 85%, 80, 85%. The g p t model is 98%, and then they noticed the doctors working with the g p t model was 83%. So, you know, they're trying to figure it out, like, well, how could this skew so hard? What happened?

I had already put it together in a few seconds where it's just like, if you were putting the assessment together yourself, you're going to be immediately confident in the assessment that you've put together because you put it together just then. If something else is putting the assessment together, you're now drawing comparison between what you came up with and then what it came up with, and you've created an adversarial environment, as opposed to you simply evaluating what it had come up with and verifying whether that was true or not. I think what happened was a lot of adversarial thinking where it was like, no. I disagree with that diagnosis that resulted in wrong answers.

Which it depends what set of doctors you're actually leveraging for this study too because, I mean, let's be honest, the god complex is a thing because it's a thing. It's not something that's fictional or made up. So, you know, the studies of that, it would be really interesting to see, you know, American doctors versus doctors in other countries to, like, see how it kind of played a part in different parts of the world. You know, I saw that headline in that study, and I was like, I know we're gonna hear about this on the show. Like, I know this is just gonna make the show.

Here's the thing that I wanna make sure also makes the show is, because I want us to put a a link or, you know, people to get access to this interview because the thing that I found extremely interesting was the entire section where they started to talk about spirituality and AI as God, and Moe immediately was like, no. Because it can exist outside of itself. And to pay attention to his thinking and belief structure around spirituality and AI was very, very enlightening. And he even referenced that he's writing a book, but not writing a book, which I found funny. He's writing the book.

He calls it alive, but he said he'll never probably publish it in book form, but maybe maybe someday, like, on a substack or something like that. This is what intrigues me because we have somebody who is spiritually minded or understands spirituality at what seems to be a pretty dang deep level from some of the stuff that he was having in this conversation down to when they were talking about, you know, putting a human in a coma before the body dies to put them in, like, a mental state of, like, being able to live multiple lives inside of, like, this virtual. It got deep. But to have somebody who has a deep level of that and a deep level of AI and conversating and merging the 2, to me, was like, okay. Now now we're having conversations that matter with lines and guardrails and goalposts that we might actually want to pay attention to.

So just again, this this interview for me was, alright. I'll continue to pay attention to you. Like, you've you've got me. You've you've reeled me in.

Man, I think so much of this. And it's with some of AI's answers right now as I as I work with Claude, I understood very quickly that it was gonna suggest things that it wanted to do for me but could not. Like, you're not connected to the things that you can't alert me when I wake up and ask for my daily schedule. But I appreciate that, and it helps me at least ideate into the future of what could be possible when there is connectivity between all these tools and, you know, we talk about agents and agentic activity and stuff like that. Like, what am I supposed to do with that extra IQ if I can't process it?

Like, this part of the world is not connected enough yet. Right? And, like, in one of the interviews, it was, like, talking about this technology overhang where for the first time in history, which maybe it's the technology part. We're definitely really good at developing products that nobody needs already. But where it's like we are coming up with use cases, like, way ahead of the time that we can use them.

Right? And I'm having that question every day right now where it's like, okay. Which part of the stack is is this AI tool gonna replace exactly? Which part of the organization? Because it's clearly doing something special.

Nobody can really, like, grasp it sometimes. So it's very easy to go back into our, you know, old planning and strategy ways. We don't know how to integrate it. That's where it's like IQ. Traditional IQ is is one thing, but then, you know, we talk about the word smart when it comes to people smarts and street smarts and EQ that has been, you know, coming up, let's say, over the past 10 years.

Like, these other forms of intelligence that aren't just like, how do you perform on this test? Like, how do you interact with people? I think that's still a big part of it. Although, I will latch on to what you're saying, Nico. Like, you described, like, every marketing planning meeting I've ever been in where it's, like, idea versus idea instead of just being collaborative.

And, I gotta wait for it to be your idea before we can move forward with the thing that, you know, we've been talking about. That's, like, so proven to be a natural human tendency as we interact with AI, and we don't have to deal with that. Like, how much of that skill set were all of my like, because you can tell when people are super good at and they have figured that out. And so their whole communication structure is like, how do I get you on board? If you could just remove that.

Let me ask you a question. Is AI actually a future fix to the issue that you just mentioned? Because you have to have a collaborative mindset when you're leveraging AI inside of your business and for the things that you're trying to do. If you don't, you won't get the best out of it. I can tell you that right now.

And so is humans leveraging AI baby steps in being able to actually have a deeper, more robust collaborative ability, first with the machine and then second with the actual humans that are on the planet with them? Because like you say, Chris, in a lot of places, that's just fundamentally flawed. Or is it just it's gonna stay flawed because it's flawed because it's the human, and that human also is not gonna be able to exist in a world or flourish in a world of AI collaborative tool mindsets.

Yeah. No. I mean, that's alright. We had to identify what benefits are we getting, and that's where it should not reduce collaboration between humans. Instead, like, if we can rely on AI for this set of things that usually has to happen now instead of spending time getting your buy in because we both trust the AI and what it's offering as far as the plan and how it's gonna be executed.

Now humans get to focus on creativity and building bigger and better instead of, like, maintenance and a instead of how we're gonna execute when at the end of the day, there's 10 different ways we could execute, and they're all gonna end up with, you know, successful results because we've already done all the work to figure out what the problem is, and we have a solution. We just can't get out of our own way to bring it to market.

Do you know that you may have just opened a whole can of worms for a lot of people viewing or listening to this? Because you it rolled off your tongue with elegance. Like, because we both trust AI, Can I see a poll on AI trust? Can we see some stats around if humans truly trust AI or if it's the exact opposite of how us nerds who are, you know, running in the front of the crowd towards, like, this thing, like, in a general human populous standpoint, if we were to be able to pull the the entire planet right now on do you trust AI? Like, how many people would be like, that's a big negatory.

That's not fair, though. Because, like

I'm not saying it's fair. I'm just saying he said we both trust AI, and we have to remember that at the end of the day, we go to work. We have to deal with humans. And so we get these mindsets and these beliefs because of who we are and how we're dealing with it and the people that are in our circle and how our circle's dealing with it. Because by the way, these 3 humans in this podcast, safe space, We're gonna generate something.

We're gonna look at it. We're gonna tweak it. We'll get it to the point where we would say, I trust this to move forward and make a decision on. I'm just saying that's not all of our listeners. That's not all of our viewers.

No. And I I think it's perfectly fine to have that divide, right, from a marketing standpoint. Think about it. Go through your grocery store. There's sugar free.

There's low carb. There's no fat. There's non GMO. There's grass fed. Those are all different businesses.

They're all competing for shelf space. So pretty soon, you're going to have human led, which I have already seen many billboards. Billboards on the road here in Ohio, of all places, advertising human led marketing, human led health care. So that market is already taking form. Why?

Because there's going to be people who are, to your point, anti AI, anti, you know, robot assistance, anti whatever it might be. You're gonna have hospitals that have AI and robots. You're gonna have human led hospitals. There may be even more human jobs as a result of this. You know, even Eric Schmidt from Google was talking about the type of job growth that's gonna happen in the future.

Hold your horses hold your phones and eat them. You know, it may be the case where a lot of us go back to artisan type work as well. There's plenty of people out there that would rather be woodworking than sitting in an office. Right?

All of them?

You can do the poll on that. No. I'm just saying that,

like Fill in the blank. You could do a poll Fill in the blank. People who like woodworking. Almost everybody in the world would rather be doing that thing in the blank than being in an office tied to a phone. And the only time I've seen it not be that way, like, they have not had help realizing the value that they could be sharing.

They've grown up in this place where this is how you do it. This is when you get a job. This is you could just gotta put in the 5 years and then maybe you could do this. It's like, no. Right?

And that's where what I was experiencing because this first happened, how I I got to this trust piece. Like, that becomes the most important piece. Surprise. Like, humans don't trust humans, and that's why we have a lot of the issues we have coming together and stuff. But it was giving me these timelines of, like, this project.

Okay. Here's your 3 months, 6 months, 12 months. And I was like, dude, it's AI time. 2020. I don't have 12 months.

And I know for a fact that you allow me to go faster than this. So get out of, like, the human tendency to, oh, we've gotta take this much time because that's how much time it takes to get everybody in the room consistently and on the same page because we gotta talk about it once a week. So then, of course, it it over cracks and it's like, alright. Day 1, day 2, day 3. I was like, okay.

I appreciate your enthusiasm to get it done that quickly. So that's where it became like, what would it take to go that fast? And we came to trust. So then every milestone became, like, how are we building trust to be able to move forward with the project, not how many meetings do we have to have or how long is it gonna take. Like, it's a completely different measure.

But once we have that trust, man, we can take the time doing the artisan stuff because the the admin stuff is just taken care of.

One of the best ways that we can start to learn about how to build that trust is to wake up with AI.

That's a wrap for this episode of wake up with AI. We hope that you feel a little more inspired, a little more informed, and a whole lot more excited about how AI can augment your life and business. Always remember that this journey is just the beginning, and that we are right here with you every step of the way. If you love today's episode, don't forget to subscribe, share, and leave a review. You can also connect with us on social media to stay updated with all things AI.

Until next time. Stay curious, stay empowered, and wake up with AI.