AI: Hype or Help? AI vs. Human Jobs with Nathan

Do you think AI is hype or help? Do we see that the AI is going to replace jobs? I mean, that's one of the biggest fears that the workforce has at the moment is the replacement of jobs. Um, I guess if you're a worker in the US, you're concerned about your job going offshore. Now, we're also concerned about our job not only going offshore, but going to AI.

That a lot of the basic data entry work will probably rightfully lose its way to the newer and newer models as they get more complex and as they can handle those edge cases.

We see this a lot with turning unstructured data into structured data um already and that has been one of the biggest use cases for AI in an enterprise or a business setting right now. What I really love about ChatGPT, it's technically meant for students, but we use it all the time. It's called study mode. I don't know if you've heard of this.

No, no, that's new. So yeah, this is fantastic where if you have study mode enabled and you ask it a question, it will not give you the answer. It will help you think through how to get to the answer.

So we are all for using ChatGPT, but I am hiring someone for their critical thinking skills.

Those were great examples of how you've nicely blended the human element with the AI element. I'm very interested in I guess virtual virtual uh workers, remote workers, time zone leveraging that. Is there some kind of AI magic tool that you could reach out and say that would really help me? Is there something that comes to mind for you?

So not on the performance management piece, but there is definitely on the efficiency piece.

People, we love connection. We crave connection. We need connection. And how about using AI to help with that? You know, I think that's a a really positive idea that comes out of AI. Good day and welcome to the show. I'm your host Pete Cooper from Skillion.

Our topic today is AI Hype or Help and I'm meeting with Nathan Sykes. Now, Nathan owns his own company called the People Company. They offer virtual assistance to small businesses to help clients optimize their time and delegate some of the work off their shoulders. Now, this is a really important topic for AI because I think a lot of people feel that virtual assistants are going to be uh taken out of the workforce because of AI. Now, Nathan has some opinions about that and we're going to find out all about it and whether AI is helping or hype. Good day, Nathan, and and welcome. So, you uh you were telling me you're a digital nomad. Um and you also founded the people company.

So, maybe we can talk a little bit about that. Um, so tell me what is the people company about in in in brief?

Sure. So the people company places excellent virtual assistance with small business owners. And I think a lot of people especially nowadays get it a little mixed up and they think a virtual assistant has something to do with tech or with AI. And that's not exactly true. Basically, a virtual assistant is a remote professional that happens to live in a country with a lower cost of living um and allows you to hire someone who speaks excellent English, has excellent competency at what they do, just at a lower rate than hiring someone in the US.

So, we help small businesses here in the United States tap into some of that excellent labor elsewhere. So, they're saving on costs and people from less fortunate countries have ways to get higher levels of employment. Oh, nice. Very nice. And what what was uh your original spark to get that to go into this business? What was motivating you?

Oh, goodness. You know, through high school and college, I put myself through school with a lot of uh tinkering on the side, right? A lot of, you know, lightweight marketing work, but more specifically, a lot of lightweight development work.

I fancied myself a little bit of a software engineer and I was building small internal tools for people and that's how I had some spending money in high school and college at least the first few years of college and I started to outsource some of that work out to other people but you know being in college I couldn't afford to hire Americans so I looked at Upwork and I started seeing that there were people in other countries who were just as talented just as skilled, spoke perfect English and I started giving them some of that work and then very quickly realized that there was more opportunity in that than there was in the development work.

So, as a sophomore in college going into my junior year, I founded the people company and I've been at it ever since with our team of almost 30 people full-time now.

That's great. Right. Yeah. So, I mean, you and I both have used Upwork and Freelancer and those, but I'm also one of your clients as well. So, I know why I use Nathan Sykes and the people company, but how do you explain to somebody else why they would use uh your company versus one of the competitors?

Small businesses are dragged in a million different directions nowadays. Both small business owners and medium-sized business owners are stuck wearing a million different hats.

And it means that a lot of their day is not spent working on the business or is spent relaxing with family or whatever they feel that they deserve as a business owner having taken that risk. Instead, they spend their day doing mindless, needless administrative stuff that either they're not good at.

It is objectively beneath them as a business owner or they don't know how to do it or in some cases they just feel like they don't want to do it which is a perfectly valid response and I think postcoid a lot of especially main street business owners which is the target demographic we service at the people company realize that you don't need to be in the office to do a lot of that work even stuff Mail handling right now could in theory be done by a virtual assistant because of virtual mailboxes and stuff, but we're very quickly expanding the scope of job responsibilities that can be done by someone who is not sitting at a desk in a brickandmortar location.

And with that, it means that not only do they not need to be in the same office, they don't necessarily need to be in the same state or the same country. So, we have been working with a lot of small business owners who have realized, hey, I'm spending three or four hours a day, checking emails and calling people and paying bills and chasing down invoices and scheduling things on social media and editing these videos or, you know, we do a lot of work with um um with insurance agents and with lawyers and there's a lot of correspondence that has to be drafted.

And every minute that a business owner is spent doing that work that could in theory be done by someone else means that they're not focusing on the actual needle moving activity that will actually grow the business. So we found a great niche in the main street business owner who has just come to this realization that a lot of this work doesn't need to be done by someone in the office and because of that can be done by someone that's not even in the same country or the same time zone.

Yeah. Yeah.

Yeah, I mean you talked a little bit about postco uh what I witnessed postco or during co even was the shift from people who were using the phone as the main means of communication to using zoom and video.

Yeah,

that that shift seemed to happen during co

I guess there's been a little bit of shift back but I think there's a there's almost it's stuck now. I think a lot of people have moved over to Zoom and to those kinds of tools. Is that what you've witnessed as well?

Yes.

I mean, for the life of me, I I still can't figure out why Skype didn't win this war because they were like right in the position to, you know, really capitalize and then Zoom came out of of of nowhere and then, you know, a Microsoft Teams, I think, eclipsed even them. So I still for the life of me don't know why Skype didn't get the the crown there. But yeah, you're absolutely correct. You know, I wasn't in this line of business when CO originally hit,

but even just anecdotally, I can see that shift. And a lot of people, you know,

our firm does a lot with trades people, right?

A lot of roofers, a lot of plumbers, electricians, appliance repair, people that you would not traditionally consider techsavvy, and they would be the first to admit that, right?

And they're having to adopt all of this great technology as a result of the COVID pandemic, forcing them to have to collaborate with their team and with their other stakeholders electronically and digitally.

So, a lot of them having lived and died from the phone and having used the same phone number for business and personal for 25 years are now having to go set up on on Zoom and they're really, you know, getting a crash course in tech 101 that was forced by this pandemic and that only feeds into what we're doing, you know.

Yes. Amazing. So, um, well, we talked a little bit about tech. Let's go a little bit deeper into that. How you would set somebody up, what tools that you generally use, how you handle things like the time zone differences between say the east coast of the US and somewhere out in Asia.

Just talk me through some of the ways you you kind of start to integrate somebody who's remote into a business that's here in the US. Yeah. So, we always say that the number one thing that really matters when you're bringing in a remote team member into an existing business is context. Context. Context. Context is the most important thing above anything else. And it is true. If you're sitting next to someone in a physical office setting, you overhear things that help inform you as to how the business runs, how it generates revenue, what its growth potential is, what its challenges are, where the skeletons are buried.

You get all of that just by being in the same environment and overhearing things, right?

And you don't get that by default when you're working with a remote team member, which means you need to foster a culture of building that in from the ground up. A great example is internally at the people company, you know, we have four head office staff that then supervise all of our client-f facing VAS, you know, the 25 or so um the 25 or so team members that we have actually paired with a client,

but we run a default open company, which means everything for those four people at the top, regardless of whether or not it affects their job responsib ibility anyone can access.

And that allows them to see how the company runs even outside of their main job responsibility

and helps inform them as to where we're going, where we're at, what our challenges are, but more importantly allows them to get all of that kind of a peripheral vision that helps them make decisions outside of the scope of a standard operating procedure or outside of the scope of their instructions. That way they're not bothering me or they're not bothering their supervisor all the time. They have enough context as to what the goals of the business are and how the business runs in order to be able to make that judgment call.

More tactically, when it comes to tools and software, we have a lot of exciting things that we help get our clients on. Because we're so white glove, we really adapt to whatever our clients are using. So, I think we're trained on like six different phone systems at this point just because no one can agree on the same thing. But our standard stack um especially when we're helping move someone like a tradesman onto uh new software to help them collaborate phone systems, we love uh Quo. Uh it is I think a formerly known as open phone and we love it because it uses a lot of these great AI features and uses it in a very smart way where all calls get automatically transcribed and summarized.

You know, you have the two-way integration into most popular CRM. Um and it allows you to keep track of what was said and what the conversations previously were, which is just fantastic. We use a lot with the Google Workspace suite and their new Gemini products which allow you to integrate the generative AI across their entire suite. So, you know, a Gmail, it can draft, it can schedule, it can parse emails uh with all of the drive functionality. It can now, which is so awesome, it can use the AI formula inside of a Google sheet. So you just type equal AI and then you can build in prompts into your sheets and then pull in data from other cells which is super handy.

So um you know we do our best to equip all of the virtual assistants that we place with a client with all of this very forwardthinking futuristicesque technology that helps them get better at their job so they can help their clients more.

That's amazing. I mean you you jumped ahead of me there. You've actually got you've actually mentioned three AI tools. It's going to be on my next question about what AI tools. So those were great examples of how you've nicely blended the human element with the AI element. But I guess the next question is do we see that you know AI is going to replace jobs.

I mean that's one of the biggest fears that the workforce has at the moment is the replacement of jobs. Um I guess if you're a worker in the US you're concerned about your job going offshore. Now we're also concerned about our job not only going offshore but going to AI.

How do you see this unfolding in the next few years?

I think that a lot of the basic data entry work will probably rightfully lose its way to the newer and newer models as they get more complex and as they can handle those edge cases. We see this a lot with turning unstructured data into structured data um already.

And that has been one of the biggest use cases for AI in an enterprise or a business setting right now. And

I actually think that that's a good thing just because that's really mind-numbing work anyway and has high turnover. It's quite soul sucking. We try not to place a lot of it and we try to take really good care of people whose job responsibility it is just because it's so repetitive and it's so menial but it it needs to be done. So stuff like that, you know, absolutely, you know, uh I think the AI can take all of that. Where you really get into more of the the the sticky wicket is more of of the generalist office manager work, right?

And I think an argument could be made that some of the job responsibilities will probably be able to be done entirely by artificial intelligence. Um what is more realistic I think is is a human being will still need to be in the loop

and that's where the virtual assistant comes in

and if we shift our focus from at least in our case you know how do we prevent our industry and our business from being entirely replaced to how do we train them on how an LLM works works. So they know how to use it intelligently and they know what its capabilities are. How do we train them on not just individual pieces of software, but how does the underlying technology work?

That way it's kind of evergreen and they know how to approach the new version of whatever the software is in six months or the competitor which all of a sudden leaprogs over the other, right?

How do we help implant those evergreen skills and how do we help our VAS think critically about artificial intelligence on a fundamental almost like a theory level

rather than helping train them in specific pieces of software because as we've seen over the last two or three years you know it is this is the high performer here and then three months later someone comes along with a better model and then you need to go learn an entirely new user interface or you know a same thing happening with a lot of the very specific use caseoriented pieces of software where you know back when the ChatGPT models built into the um the chatgbt UI weren't as good as some of the marketing generation right

you saw one software after another that was tailor made for marketing use and it was like every month and a half or every two months you would get a new software within even flash your demo and even better performance and then you'd have to move, you know, um all of your context and everything to that new software because it was like the cutting edge thing. And it was like this almost I mean almost this race to the bottom, you know.

And I think that a lot more could be said about helping train and educate people about how AI works, how um how safety and security uh which is a big thing that we train about internally, you know, not putting client data just straight into ChatGPT.

Um and the implications of how that data is used to then train further iterations of the model. You know, all of that stuff that nobody really talks about, which will help equip them to use it in a way that actually um actually helps benefit the client. That's a very long way of saying my gut feeling right now having been in the industry is we're not going to see a complete shift to where you know the role of the virtual assistant or the role of a BPL company, a business process outsourcing company is negated entirely.

Rather I think if we do the leg work and we train them on how to use these tools and these models and this underlying technology in a responsible way that's going to be the play going forward

right so here I could think of a couple of examples to illustrate that one um might be you wanted to someone to design some uh a graphic bit of graphical art like a logo or something going back before AI you would have to find somebody perhaps through one of those tools, brief them, and then ask them to give you three or four, you know, choices. Um, and the better way you've created that context, the better way you get those three or four logos.

Yeah.

Now, of course, I can go to an LLM, give it a good prompt, and I can get those three logos autogenerated. So that's kind of one example where I think that traditional I guess scripted kind of role is going to be diminished by the power of AI.

But then the flip side to that is something I think you're bringing in where you've added some value over and above and on top of if I'm understanding this correctly over and above and on top of these freelancer and Upwork type tools that actually brings almost an education component in to people who uh aren't aware of how to leverage virtual assistants and AI and uh you know to streamline their business to be far more efficient and take the load off the business owner. I think that that's a great way of putting it.

Also, I don't know if if you've seen this, but you know, being on the younger side of things at my age, my compatriots, my my colleagues that are not in the tech world,

what I've seen is almost a contempt for AI. And I have noticed that there's a lot of conversation, there's a lot of discourse going around about, you know, the AI's effect on the environment and on water usage.

I'm not going to speak on on how accurate that is. That's not my my place. But there will always also be a market of people that in the role of a business owner have have no interest in at this point engaging with uh the technology

and they will need human support.

But then the flip side of it is, you know, we have a VA placed with a massage therapist

right now and and it's a part-time engagement. You know, it's a 20 hours a week and that means that the phones are only covered for, you know, a four hours a day, right?

And they're open 10 hours a day, so they need some coverage on the other six hours that they're open as as well as after hours, right? And we originally placed them with Quo and that's our, you know, a very a forwardthinking um the phone system that we're a partner of

and Quo has this built-in AI answering service actually one of the best on the market that I've seen uh called Sona. Mhm.

And even though it's the best on the market, even though it is incredibly lielike, incredibly realistic, super low latency, I think we got 6 weeks into using it where instead of of them continuing to use it, um they just shifted the part-time hire to full-time because, and I quote, "The people of Redmond are sick of AI was the specific response that. " So my my um the other side of it is you know it it it could be that people are are already missing the human touch.

You know, I know that's why a lot of people work with us is because a a live human being always answers the phone within two or three minutes and you can get in touch with us very easily and and you know that there's the human element there. And just anecdotally, it is funny to me that a lot of people are already getting kind of sick and tired of having to interact with these robots and are already willing to pay a premium, you know, more than double what they were paying before for a part-time hire just to avoid it, you know. So, just with those like anecdotal stories, I think that our industry is safe uh for the time being.

So our attention then turns to well then how do we empower our virtual assistants to do something good with the technology you know

right? So that's a good pivot point. Let's let's talk a little bit about how you guys uh use AI and one of the challenges that I've had with with especially my new hires and people that come perhaps just straight out of college fairly early into the workforce. Um, there is a tendency for them to to uh take what I've said, a question, and go straight to the LLM, uh, type in what they thought they heard,

uh, and then send it back to me as if is this okay? And I I can tell because I can tell that it's it's too uh too refined. It's too it's too quick.

And of course, they can't really explain um what it is they've delivered to me. And then I'll explain what I want again. And then I think they'll go straight back to the LLM and try again. And it's almost as though they're taking themselves out of the thinking part um and just going straight to the LLM. You I guess first question then is are your virtual assistants using um LLMs and and if so how and how are they using it in in good ways?

Yeah, we see exactly what you've described a lot with our test tasks and these are paid assignments that we do before we place a virtual assistant with a client and that's one of the reasons why we do it because it is getting significantly easier the more that we see it to be able to read what people call a ChatGPT syntax right um you know you're you're able to tell it's getting easier you know the m dashes the it's not x it's y with like the hyperolous, you know, a second example there. So, it is getting easier and especially if English is your first language and we're hiring from a talent pool where English is not their first language. It's usually their second language.

We can usually tell when they can't. So, we then, you know, have a bit of a dilemma. Is it, you know, is this as good as it's going to get or is there some, you know, a breaking in that we can do with this person? And you know set expectations appropriately that you know hey our client is not hiring you to like run ChatGPT like you know if they wanted to run their question through Chad GPT they wouldn't be spending you know a $15 an hour to hire you right um I they could just do it themselves so we do heavily encourage that our VAS use LLMs we're a big ChatGPT that's the model of choice um you Claude and Grock are all fine choices.

We just really like, you know, a ChatGPT a lot of people are very a familiar. Um, and what I really love about ChatGPT, it's technically meant for students, but we use it all the time. It's called study mode. I don't know if you've heard of this.

No, no, that's new.

So, yeah, this is fantastic where if you have study mode enabled and you ask it a question, it will not give you the answer. It will help you think through how to get to the answer.

So, we are all for using ChatGPT, but I'm hiring someone for their critical thinking skills, and I'm all for refining that.

So, it is one thing if you use study mode and you put the question in there and then you communicate with the system in a way that helps you arrive at the question because then there's some upskilling happening there, right? Then it turns into me investing in your growth as a team member with the hope that you know I retain you for a long time and then you you know learn more and then you know your capacity for critical thinking and your intuition um you know all get better over time. That's completely different from you know what you were describing which is you know a putting my question into an LLM and then basically just copying and pasting it right back out.

Yeah. Yeah. Yeah.

So, I found that uh uh I guess a little frustrating when I see that happening. And this is a a really neat tool to try and uh to try and get around that. Um and I think it also what I'm trying to do I think is the same way as you Nathan is to yes use the AI tools but don't take yourself out of the thinking loop. What I'm trying to do is develop um uh someone who has as you say context that grows over time. They they know uh how to take the business forward. They're they're they're on the front foot. And we have a leadership tool that we use. It takes us through four phases of leadership. And I'm always trying to get them to stage four, which is effectively um they're running the project.

And I'm just sort of um at a distance asking them what they need from me to help prod that forward. And I can't get them to that level if they are just relying on uh ChatGPT reports and me giving them feedback on on ChatGPT's reports. They're almost out of the loop. You know, they'll never get to the bit the point that I need them at, which is to drive

projects and ultimately take off take some of my load off me

so I can fully delegate. You know, delegates what I'm trying to do and trying to always trying to delegate, push things to people so that I don't have to do them.

Not because I'm lazy, it's just that I've got another one that I'm going to bring down, another project I'm going to bring down as soon as I can get that one off. Um, never ending.

I think a lot of people who are doing what you're describing are not thinking things through to the end because sure, they probably can get away with that. And I'm sure, you know, a lot of people can take the output, they can, you know, remove the m dashes or whatever they've got to do to make it seem like they typed it out and, you know, make it look like it's it's their work.

But I mean, if that's the end goal and if that's what they're doing, there are millions of people that are doing the same thing, right? That is a dime a dozen at this point. You know, we get 250 300 applications for a single role at this point here and that's a lot of people. I I I quite literally mean, you know, a dime a dozen in the sense of I have 250 people that can do exactly that. But what's going to happen is that people who can think critically and who can absorb context and who put a little more effort in, they will become in lesser and lesser supply and in higher and higher demand.

And you know economically they'll be able to charge significantly more competitive rates because just for doing what we consider the bare minimum, the rest of the workforce is not even going to be able to do that. And then you know I think over the long term they'll find that it's a significant advantage. So that's

so AI could effectively dumb down an element of the workforce who is not really engaged in uh solving the problems themselves but relying you know so much on AI.

I mean, I just literally signed up for my guys to use a coding tool, reluctantly a little bit to be honest, even though, you know, we're we're we're so pro AI at Skillion, of course, but um I had to sit down with the guy and say, "You have to know every single line of code.

Use the tool, but um don't rely on it, you know, because often we found ourselves getting into projects where we get stuck.

Uh you know nine ten of the way through customer wants it delivered and it we can't get it to work because there's so many things that have been done by AI no one actually understands how it's being created so very very aligned with you there um Nathan so back to the people side of things is is there some ways that you're able to uh gauge the performance of people uh remotely from a different country, different time zone, um or you or step away, you know, working directly with them. And I guess does AI help you in any way to gauge the performance of your uh your your your staff, you know, your virtual assistants?

So, we do have a lot of the performance management side very well built out, but it's actually all I wouldn't say um old school because it's all a proprietary software that we've built from the ground up, but there's no AI in any of it actually. So, we do uh performance scorecards, we have an end of day check-in, we do individual coaching sessions. We have that very methodically built out. And our clients don't even see 90% of that because it's not their burden to bear. Their expectation is that they have someone that they can rely on that's on top of their game. They don't need to see any of what we do to help the virtual assistant be their best self, right?

Yeah,

but it it does exist and we've built a lot of software and a lot of technology to help support our office since we are only four people, right? So, we have a lot of that standardized and a lot of that is, you know, very polished, but no AI in the performance management side of things,

right?

Because, you know, at the end of the day, we're dealing with human beings and it's it is an art and a science, right?

Right.

So perhaps you're talking about interviews uh with the client, interviews with the virtual assistant um that can help you with the scorecard and maybe there was some written um guidelines from the client that you can say well how we hitting those goals that that type of thing.

Yeah. Right. Cool. All right. So if there was some kind of AI tool that would help you with that. I'm very interested, I guess, virtual virtual uh workers, remote workers, time zone, leveraging that. Is there some kind of AI magic tool that you could reach out and say that would really help me? Is there something that comes to mind for you?

So, not on the performance management piece, but there is definitely on the efficiency piece. So, I I am of the belief that, you know, we're dealing with the human beings and this is their livelihood. It is in many cases their sole source of income.

I feel like when we're judging their performance and when we're making decisions that may affect in their long-term uh prosperity and their long-term you know economic health and

I mean if they stay with the company or not or if the client is satisfied with them or not those decisions should be made by human beings you know a computer can never be held accountable therefore a computer should never make a management decision right or in this case a a decision that can affect the livelihood of a real life human being. But efficiencywise, absolutely. One of the things that I really wish existed. Every single piece of software everyone uses right now has hot keys, right?

These are are a specific combination of keys that you can hit on your keyboard in order to navigate between different workflows or to autofill information or to move between different panes. Every single software, it doesn't matter which one, has some sort of a keyboard-based support.

And what we see a lot right now is that our virtual assistants or really anybody I mean even our clients don't take the time to learn the hotkey for a specific action like we see this a lot on Google Chrome right um you know you can move tabs on Google Chrome by hitting option command and then your right or your left arrow and your hands are already at your keyboard so it turns you know having to change the tab from like having to move your mouse click on the tab move it back to where you that takes, you know, a 3/4 of a second. Doesn't sound like a lot, but it really adds up to where if you just memorize the hotkey, it takes a tenth of a second, a 20th of a second.

And that compounded over,

you know, 70, 80 actions that have hotkeys adds up to very realistically 30, 40 minutes a day of you not having to go and click specific things. And I really wish that there was this AI powered tool that you know almost predicted what a user was going to do and reminded them that there was a hotkey for that. That way they can get more efficient at navigating the software that they use on a daily basis.

That's amazing. Amazing idea. Maybe that's something we should look into for you.

It would save everyone a lot of time.

Yeah. Okay. So um I guess we're coming back to sort of wrapping up now and and there's you know the future of AI.

I mean where do you think the future of AI is going to take us and particularly in your field that's cast forward say about five years. Where do you think we're going to be with AI?

I feel like you're asking the wrong person. You know, I think one thing that we have started to use AI for that I think we'll be able to take to more of a natural conclusion is building a personalized experiences for people

at scale and and and specifically in a work context. I mean that we usually by default a record everything. So all of our emails, all of our meetings, all of our phone calls all feed into HubSpot.

And we have right now a workflow where if someone says a personal piece of information, like if offhand they mention, you know, oh, I'm sorry my late I was dropping my son off at soccer practice. That gets isolated and gets added to a custom property that we use just to keep track of of of a of of this like you know anecdotal um almost a personal information about someone's life that we can then use to feel more connected more personalized to them.

So, like right now I am actually uh Christmas shopping for all of our clients and we're using those little snippets of info that we've gotten through all of these uh conversations um to, you know, help help us influence what gifts we're going to get for people. And

it helps us a lot because, you know, we had a prospect

um 3 months ago off-handedly mention that he's allergic to meat um to chicken, right?

And we do these lovely customer appreciation gifts uh through a gold belly and um you know, I use that and I was able to make sure that we got him a vegetarian, you know, offer instead.

So, we use stuff like that already, but I think that there's a lot of room for opportunity there where, you know, there's just so much personal information about everyone we interact with that's just dropped in these little anecdotal 5-second, you know, interactions that you'll forget unless they're written down somewhere. And I think in in five years we'll be able to see that kind of get to its natural conclusion where you know every time we're on the phone together or every time you as a customer interact with me as a vendor we're able to hyperpersonalize it even though there's really no need to. It just makes you feel better about doing business with us.

So I'm really excited about those kinds of opportunities. I feel like the act of remembering someone really makes them feel good. And I think the AI can really be a great force for change there.

Wow, that's amazing. I think you've really touched on something very important and that is that as people, we love connection. We crave connection. We need connection. And how about using AI to help with that? You know, I think that's a a really positive idea that comes out of AI. So to summarize, Nathan, do you think AI is hype or help? Help.

Yes,

absolutely.

Good.

Help, help, help.

Well, that's been great. You've educated me and our audience enormously, uh, Nathan.

So, thank you very much. Um, and I wish you well with your business and, uh, take care.

It is my privilege and thank you so much.

Well, that was a fascinating discussion I had with Nathan. We really learned a lot about how virtual assistants work and how AI is being integrated rather than replacing virtual assistants. Um, we talked about how LLMs are being used by virtual assistants uh to help them perform better. And Nathan had some really interesting insights about how AI can actually help us with one very important human condition and that is our need for connection.

I thought that was a really good insight and how AI can actually be helpful and how it's going to drive us forward to do things and be a really useful tool. So, if you want to know more about Skillion, please follow the links in the description. If you want to make a comment about this podcast, please do so. We always like reading comments from our audience and direct us into things that are interesting subjects. We're all about AI, AI hype or help. I hope you found that interesting and thanks for watching.