AI Value Track Podcast - Episode 2

We dig into practical AI that saves time: personal LLMs, embedded features already in your tools, and AI‑native agents that take action, not just give answers. We share proof points, pitfalls, and simple ways teams can adopt safely and quickly.

• Personal LLMs as thinking partners and draft accelerators
• Embedded AI in email, slides, CRM, and meetings
• AI‑native tools mapped to roles like SDR and legal
• Agents that act: scrape, configure, reset, and schedule
• Quality, speed, and evidence versus laziness concerns
• Training, culture, and best‑case sharing for adoption
• Guardrails, verification, and human‑in‑the‑loop control
• Real‑world cases in support, retail, facilities, and QSR

#theproductivityexperts
Register for the 2026 Productivity Forum
Find us in the Top 50 Productivity Podcasts
Connect to Simon on LinkedIn
Follow ReThink on LinkedIn

Chapters

• 0:00 Setting The Mission: Practical AI
• 0:54 Personal LLMs And How To Use Them
• 2:26 Embedded AI Inside Everyday Tools
• 5:04 AI-Native Tools And Job Redesign
• 8:49 Evidence, Risks, And Human In The Loop
• 12:34 Adoption Tactics: Training And Culture
• 16:24 Agents 101: From Prompts To Actions
• 20:34 Scheduling, Automation, And Orchestration
• 24:14 Real-World Use Cases And Quality Loops
• 30:19 Closing Thoughts And What’s Next

Transcript

SPEAKER_02: 0:06

Welcome to the AI Value Track. This is podcast two in a series of three. I am delighted to be joined by a colleague of mine, James Boll, head of Data and Insights at Rethink Productivity. Hi, James. Hello, Simon. Thank you for having me. And returning from episode one, Ian Hogg. Thanks, Simon. So we're going to get into the nuts and bolts today. We're going to talk about those tools maybe that individuals use in their day-to-day lives in business to make them more productive. You know, we talked about the benefits in episode one, freeing up time, all that kind of stuff. So we'll jump straight in then, Ian. Talk us through some of the tools that are available, you know, for individuals at corporate level to buy that they might be using now or they might want to start to think about using.

SPEAKER_00: 0:50

Fine. So there's, you know, so the probably the first place to start is is the sort of very personal tools like ChatGBT and other they're called large language models, but they're basically LLMs. LLMs. That's the way.

SPEAKER_02: 1:03

I'll I'll include you you spell it out and I'll introduce them.

SPEAKER_00: 1:05

Yeah, so they're called LLMs and then the and they are they're they're a sort of you you like a chatbot where you talk to it and you type into it or talk to it verbally and it gives you an answer. Okay.

SPEAKER_02: 1:19

And they're they're all sophisticated, aren't they? With is it a quick check? Is it I think one of them's got a deep reasoning where they go off, one of them will write you a deep research report. I was using one the other day, maybe Gemini, and it said, Do you want me to build you an infographic of it at the end? And it was like, yeah, if you really want to, it'll it look great.

SPEAKER_00: 1:35

But exactly. And they're the sort of things where you give it an individual, you you write it as what's called a prompt, and you're giving it a task, the the model, and it comes back with an answer. And that answer could be a PowerPoint presentation or it could build you a business spreadsheet.

SPEAKER_02: 1:49

Again, my naivety, but I'll say it. You can't treat it like Google, right? So you can't well you can, but you'll you'll limit use. You can't say, find me the best laptop for business use under a thousand pounds, because that that's really what you use Google for, right?

SPEAKER_00: 2:02

Well, here's the difference is if you asked a a good LLM, what it'd say, well, tell me a bit more about what you want, it would come back and ask you some questions. Trying to refine the search or it would say, What do you want the laptop for? You know, what what's your budget? You know, that it would come back and ask this question. So that's like the first sort of area of these personal tools. The second one is what what I what I would call sort of embedded AI in existing tools. So if you think about it, most companies have got tools that everybody uses.

SPEAKER_02: 2:31

So Gmail, yeah, so yeah, email, iPhone, iPhones have got the iPhone, whatever it is now, 16's got their AI in, hasn't it?

SPEAKER_00: 2:38

Yeah. Yeah, but what I'm saying is there's software tools that the whole company uses. So it could be your email suite, it could be your your video conferencing suite, it could be something like Jira, which is like a ticketing system for problems, or project management software, or CRM. So all of those sort of tools. So what I would call embedded AI in that is the sort of thing where you make available, for instance, note takers within your you know, if we're doing a video conference, you you everything always wants to take notes.

SPEAKER_02: 3:11

Everything summarizes it and then send me an email with it afterwards.

SPEAKER_00: 3:15

Exactly. But you know, the that that sort of stuff. You might have you might have a if you're writing a ticket, like a problem ticket for to send to a help desk, it might prompt you with a format and then come back to you and say, You haven't answered this question. So to that, I would I would call that embedded AI into existing systems, and those can can help productivity. We talked in the first episode about the like scheduling, you know, and scheduling assist. That's a good example where you've got a workforce management platform as an example, and then you've got a tool that helps you do the even in when I'm writing a PowerPoint now, the little pane will come up and say, Do you want a different design?

SPEAKER_02: 3:57

or Do you want me to, you know, it almost auto-corrects or restructures the sentences at times now rather than just chat my appalling spelling.

SPEAKER_00: 4:04

Yeah, yeah. And and those are and some of those have got the LLM embedded into the system.

SPEAKER_02: 4:10

And then the next, just to stop on that point, so we're probably using more AI features than we actually know. Yes. Because it might not be called AI, there's you know, Microsoft used Copilot as a word, but there's the load more stuff out there than we probably take credit for we're using.

SPEAKER_00: 4:27

I and I I think now almost every software you're using is gonna have some sort of AI. If you think about it, if you were doing if you were doing a tender for a new software tool for your company, you're gonna ask the question of the sort of the vendor, you know, what's your AI roadmap? What AI do you have now? Is there a chatbot? You know, do I have to actually read the help files or can I just create just question it and say, hey software, can you tell me how to how to set this up in in your software? So AI ought to be embedded in those sort of tools. And then then the sort of third category. So the first one was the the sort of LLMs, the large language modules. The second one's embedded AI, and the third one are what I'd call AI native tools. Okay, so they tend to be, these are being sold now as sort of almost like with a they've got a job title with them. So for instance, like legal AI, you know, so it's like your lawyer in-house will do all your contracts. It is AI sales development rep.

SPEAKER_02: 5:30

So it's got a specific job title almost that then leads to the task.

SPEAKER_00: 5:35

Yeah. So so a good one, sales development rep. So you know, in in software and a lot of other businesses, that's the person we would have called when when I was younger, they would have called that tele-sales. Yeah. But now it's all on LinkedIn and messaging. And yeah, there is somebody's or several people who built an AI that just manages that sort of process. So those those aren't just sort of layering on a or embedding a bit of AI into an existing platform, those are saying, right, from you know, from a clean sheet of paper, how do I give somebody a tool that is effective is is almost equivalent to replacing a full-time person. And it and again, we talked about it in the the first the first podcast on the series is that's the sort of tool that is replacing entry-level jobs. So a senior salesperson gets his AI SDR. Not so long ago, a senior salesperson would have had a real SDR. You know, you've still got the senior salesperson, but a lot of the legwork on LinkedIn and email notifications done by an AI.

SPEAKER_01: 6:38

What's SDR, sorry?

SPEAKER_00: 6:39

Sales development rep. Telesales. Yeah.

SPEAKER_01: 6:42

In the old world. In the old world. And no telephones. But it's like I I think I might have used this example before, our productivity forum. Like people talk about robots. As soon as the robot is useful in doing a specific job, it's not called a robot anymore. You have several robots in your house, they do your dishes for you, they clean your clothes, they're called dishwashers and washing machines. AI is exactly the same. People talk about this big idea of AI, but as soon as it's actually doing something useful, it just disappears into the tools and disappears into our day-to-day jobs, and they don't talk about it being AI anymore.

SPEAKER_02: 7:13

Yeah, and again, people watching, listening, I'd challenge them to think about all those product suites we just talked about. As a business rethink, we use Google, Gmail, and we get Gemini in there, but it's also now appearing, it will give you three suggested responses to an email. Yeah. You have to be really careful because there might not be your tone or language, and sometimes you might be saying that's a great idea when the email might not be such a great email, and vice versa. So as ever with this. And sometimes it takes longer to read the three responses, and it does just uh lie. There is that. And it and there's you know some interesting stuff of it, it'll rewrite it, it'll soften it, it'll make it more formal, it'll so there's some really cool stuff that has a real impact on people and probably gets you out of some situations or helps you avoid them. But my point is whenever I personally use those, I don't think it probably doesn't register with me. I'm using AI today. It's just the way it and all those features come and new ones come, and on Google Meets now you can have different backgrounds and you can put stupid faces on, and yeah, it takes notes for you and it emails them to you, and it gives you a summary of it, and it gives the action points with whose it's been assigned to. And it all it is is a pop-up that comes up and says now you can do this and you go with it, or you don't. It it's that ingrained in what we do. I suppose the point I'm making is it's evolving quickly, but I never sit back and think I've used a lot of AI today. It's just the way it is.

SPEAKER_01: 8:40

No, and it's just it's just tools that make you more productive. So I mean people wouldn't use them if they didn't make their jobs easier, and they have names and they don't think I'm using AI to do this, they just I'm using this feature in this tool, and it's making things better. And uh people using Chat GPT to help them write tend to write more quickly, but also higher quality. So it increases increases their productivity. People in knowledge economy jobs can use large language models to make their writing more effective. So there is evidence.

SPEAKER_02: 9:11

Not just critical.

SPEAKER_01: 9:13

I mean, there's there's studies that that show that it show that it is. People can code more effectively or more elegantly using using a large language model than just doing it from scratch.

SPEAKER_02: 9:23

My ke my counter would be, you know, if I'm playing devil's advocate, are we not just getting lazy then? Well, you I feel lazy sometimes with a canned response. It feels a it feels it's in it's the right answer, but all I've done is press a button.

SPEAKER_00: 9:34

I listened to a podcast with Stephen Bartlett the other day, and it and it and they had a a guest on that was saying that by using all these large language models we were all going to get dementia because we were so lazy that we actually weren't thinking for ourselves. So possibly I'm I'm no expert on that one.

SPEAKER_01: 9:49

So I'm possibly but you've you've got, you know, I I read a case study where T-Mobile had used some AI to, while an agent was on the phone to a customer, the AI would read through that customer's service history and how many times they'd called and what they'd called about and would prompt the agent with questions to ask the customer. And guess what? Customer satisfaction went up, job satisfaction went up because they were having nicer conversations and sales went up. And so, you know, it's a win-win-win. If you're using the tools, I think you talked Ian in episode one about hu be having humans in the loop and using AI to enhance people, you can do this not to make people lazy, but actually to make them better and enjoy their work more. And that's that's what you're looking for. And I think, you know, there probably are people that use it in a lazy way, like the lawyer that put all their case details into ChatGPT and in court cited a case that ChatGPT had made up and was called upon it. And that's the danger, right? And that's the danger.

SPEAKER_02: 10:45

Exams, kids doing uh coursework on them, that there's a whole while ever there's a positive side to this, there's always dark sides wrong with it. There's there's other ways people will use it.

SPEAKER_01: 10:55

Yeah, but if you engage your if you engage your brain, if in your organization you're giving people the tools and showing them how to use them properly and how to think about them, then I think there's lots of evidence that productivity is increased. And also, I mean what's interesting is there's a Deloitte study looking at when you talk to leaders about how AI has changed their organization. I think more of them still talk about it's saving us money than it's giving us new ideas and making the quality better. But it's it's proven in tests on individuals that if you are using the tools in the right way, then it can make you more productive.

SPEAKER_00: 11:27

I I also think it's to the going back to the laziness point, Simon, is really a lot of the the tasks that it's taking off of people are the boring, mundane stuff they don't want to be doing. It's like you know, looking for a bug. So if you take some software developers, it's really good at like this doesn't work. I've been yeah, I've been playing around this, I can't find, I can't crack the the puzzle. It looks at you and says, Yeah, you've got a comma missing there. You know, it's like yeah, and wood for the trees, that kind of stuff. So and then you know, we've uh we've recently trained in our uh operations team at Shopworks how to use some agents, agents. And one of the guys sort of came up with a use case where he had the most mundane horrible job uh on the back-end system that he managed to get the agent to build, and he was just delighted, you know. It's like anymore. One of those dread doing that leave it till leave it till Friday job. Yeah, the satisfaction of getting that that solved was was great for him, you know. So and I was gonna ask how do we get AI usage up?

SPEAKER_02: 12:29

I kind of feel like we're using it after the start of this podcast, we're using it more than we think. But if you still don't think you're using enough of it, what kind of things can people be doing? We talked in episode one about paid subscriptions, so making sure people've got the right the right tools and it's in that business context environment, they're not you know, maybe using their free version on their phone. But are there other things, Ian, that people could be doing?

SPEAKER_00: 12:53

So I I can quote an example we did too. One of the projects we're doing at ShopWorks is as well as trying to get our the software we sell to be much more AI enabled to use the category we talked about earlier, is I'm trying to get our team to use AI as much as possible so they can be as efficient and we can deliver projects quicker and it better for the customer. Okay, and one of the things we did is we got the operations team, so they're the people that sort of help implement new software. So there's project management, managed service of software, configuration, that sort of stuff. So we bought them all a chat GPT license, and the first thing we did was get on a Zoom call with it, and we you we asked some questions about what people knew and what they didn't, and a lot of people had some knowledge, but they they'd only had the knowledge one you know, these software tools, if you if you use them, they they exp they add new features every week.

SPEAKER_02: 13:49

So if you every day at the moment, exactly if you haven't been on it, it bumps up, goes you can do this, exactly.

SPEAKER_00: 13:54

If you haven't been on it for three weeks, you might not know what one of those buttons does. So we went to it line by line, and then one of the things we did was then you know we set sort of a task for people to go away and experiment and see if they can then come back and share their best case. For you know, that was part of the training session. So none of that's rocket science, that is that is just training, yeah. But you know, it does surprise me how few people are getting formal training on chat GPT. Of course, the best place to get training on chat GPT or any of these languages is ask it to train you, um, and it will, but even getting into that mindset is you know. So, how do you get usage up? So training, encourage people to think differently. So ask encourage people to ask ChatGPT.

SPEAKER_02: 14:43

Is it like like your colleague example? Is it almost brainstorming with the team? You know, what's all the mundane stuff, what's all the stuff that takes time? So proportionally, we spend a lot of data entry time, duplication, whatever. Almost get those sticking out, so however you want to do it, user cases to then try and use it to solve. So back to one, and maybe what we'll talk a bit about in podcast three of what other business challenges that I'm trying to solve.

SPEAKER_01: 15:10

100%. You have to read right to left when you're thinking about AI and start with the problem you're trying to solve because too much of this tech is just tech looking for.

SPEAKER_02: 15:18

Go down a rabbit hole of back to pretty pictures of cats on the moon. Yeah, yeah, you end up in all the nice stuff when you come to the end of it, you go, actually, what benefit has that had?

SPEAKER_00: 15:28

And and yeah, another one is share success stories is you know, when somebody's you know within a team has used it to do something clever. But I you know, I think it's almost an imagination challenge for people. So these things are incredibly powerful, they can do a lot of tasks. But you know, the first time you start using it, you sort of use it once you think that's clever, but you actually don't do anything productive.

SPEAKER_02: 15:51

You've touched the iceberg though at that point.

SPEAKER_00: 15:53

That's clever, but yeah, it's clever, but actually it didn't save me any time. And actually, I've now got a picture of a cat dancing on the moon, you know.

SPEAKER_02: 16:00

Lottery numbers with that was surely a way to go. Predicting lottery numbers must be that's the one that no one's talked about that, have they? Right. Can we edit that out because we're giving away IP now? I mean intellectual property, by the way, that was so.

SPEAKER_01: 16:13

I think I think the the example that I was that I heard, and I know the pace of change has uh changed over time, has increased, but after the introduction of electricity, it took about 30 years in the Western world for steam and for horses to be replaced. And people have effective ways of doing things today without AI. There's no reason for them to change them unless they see it or they they they feel it. Or dislike it that much. Or dislike it that much. And some people might dislike it but be afraid to change because they don't want to lose their job. So there's, you know, the organization has a lot of work to do, I think, in order to encourage people to experiment, to make them feel safe to experiment and and adopt these things. And it's a lot easier in what you would call digital first businesses or digital native businesses. You know, an organization that's been going 15, 20 years started, you know, when I first started my first job, we had one computer in the office that was connected to the internet. We used to fax everything okay.

SPEAKER_02: 17:05

Dial-up.

SPEAKER_01: 17:05

Yeah.

SPEAKER_02: 17:06

For those that don't know what dial-up is, that's when you had a you almost had a phone on a modem. A modem, yeah. Well, you might have a router now, similar, and it used to dial. And when it was when it was on, you couldn't use the phone to start with, could you? And you were paying like 30p a minute or something stupid, and it was like really slow. Well, ChatGPC gave me a history of the internet. Um there you go. For anybody listening or watching, that's your your first prompt. What was your prompt? Tell me about the history of the internet.

SPEAKER_01: 17:32

There you go. But that's but you know, there are legacy businesses that have got lots of data or lots of processes that exist from a time even before email and the internet, and then trying to get AI into that is clearly going to be difficult. And you have all these startups that are coming in and just doing everything with AI, that gives them an agility that's a bit of an advantage over the the big organizations. And from the research I've read, you look at China, you look at India, you look at some parts of America, businesses there are just doing everything with AI and outcompeting the legacy businesses really quickly. So to a certain extent, I think it it's important, it's imperative almost for managers and leaders in organizations to start getting people to adopt it, because otherwise you'll get you'll get out outrun, outmaneuvered by somebody more agile.

SPEAKER_02: 18:18

Yeah. And look, just to get back to agency, so where should firms begin? And we'll talk in the next bit maybe about how we get started with agents, but where should firms begin? So what things should they start to think about in terms of workflows?

SPEAKER_00: 18:33

Is it worth just clarifying what an agent is? It is, indeed. So so if you think the we talked about the the large language model, which you know, Gemini, ChatGPT, Claude, a few others. You sort of run an individual task on it. As a as a person, I log on to it, I type into it a prompt, and it does the task and it gives me the answer back. Okay, so it's a sort of a one-to-one.

SPEAKER_02: 18:57

Well that could be a recurring task, it could be a one-off.

SPEAKER_00: 19:00

On the whole, you could they don't do scheduling these things. So you could you can't, so for instance, you can't say do this task every Saturday at nine o'clock. It's you do the task and it requires you to be there to manage that prompt and sometimes answer questions. An agent is is effectively an AI normally using exactly the same technology like a as ChatGPT or or Gemini, where it is going off and doing the task for you. So here's here's a couple of good examples. So within ChatGPT, there's an agent mode, okay, and so I have a task where I have to go and gather some competitive information from five websites, there's public information, and I set the prompt going, I use the same prompt, and I tell it to go off to these five sites, it'll and it literally opens a little window, opens the sites, scrapes the information, puts it together and puts it in a report, and which it then makes available for me. Okay. Another task uh an agent might do is we use it for some software configuration. So it goes on to the settings of for customer, we've we've got these sort of prompts written, so it goes on and it might configure, say, for instance, you know, a lunch break rule on workforce management. And you you give it the standard template, it goes off, it logs on. You have to log on for it actually. It then configures it, it might, but it's more likely to do 20 of these. It's not gonna do one rule, it's gonna do all of the rules. Then it goes and creates a shift that meets the criteria for the rule and then confirms that it's correctly set up, okay, and and it can go and run that. So that's what an agent would do. Okay, so it's a specific task or a specific set of instructions, yeah. And the the productivity gain for the for the for the company is is you know, let people use their imagination and go and find out all of these individual micro tasks. So that particular task and hundreds of other ones like it, like I have a job where I have to go get information for five competitive websites. We're not gonna get a project team together, define, you know, this is the power of it. Um, we're not gonna get a project team set a budget, set some KPIs and some measurements. We're gonna just give somebody a tool, and when they get fed up with doing this boring job on a Monday morning before they can do the rest of their job, they're gonna they're gonna train an agent and they're just gonna run that every Monday morning, then they're gonna open another window and then run another agent to go and do their next task. So that that's the difference between an agent and uh and just as a standard prompt, and there's one other sort of nuance on that is those agents within Chat GPT or Gemini, they you still can't schedule them, okay. I'm sure it's coming. So it sounds like an obvious next step of every Monday Polish report for me. But there are other tools out there, automation tools, which basically can connect to OpenAI or or Gemini or Claude or any of these models, go and perform the task, and they can be scheduled. So instead of me having to wake up on a Monday morning at nine o'clock and cut and paste that prompt to go to those five websites and set the agent running, there's a tool that just wakes the agent up and does it every Monday morning. So, you know, the productivity gains are we we build we you know, each individual builds a suite of agents to automate the boring, mundane tasks. And when they wake up on a Monday morning, it's already there presented, and actually they start adding value by reviewing the information and making decisions based on it.

SPEAKER_02: 22:48

Any experience with agents, James, that it would be good to share?

SPEAKER_01: 22:51

Nothing that has yielded anything kind of concrete, but I I think the the way that agents are going, you know, everyone talks about agencai AI, and what Ian's just described is is basically what that what that means. AI is no longer it's no longer sufficient to have an interface you ask questions to and it gives you answers. You want to it's goal-based now. So you want to say, this is what I'm trying to achieve, and can your agent achieve that for you?

SPEAKER_02: 23:14

And I assume in the future, maybe the scheduling of it each week produce me this report feels like a you know, probably in this day and age a fairly simple next step. It feels like then the next evolution from that becomes, you know, scrape these five websites, produce me this report in this format, whatever, present it to the email it to these people, whatever, but also look for wider opportunities in other websites or across the web that I've missed. So what fill in my blind spots.

SPEAKER_01: 23:40

Yeah. Or I mean autonomous agents, I I guess are probably the the goal where it sees, oh, you've had a meeting come in your diary next Friday, you're going to need this information, here's the report ready for you, or you ran this report, I thought you'd be interested in this as well.

SPEAKER_02: 23:53

And there's a so that's a personal one. I also naively probably see a world where there's business to business ones and business to customer ones. So at some point I get an email text, whatever it might be, saying um Simon Edo, your personal agent has negotiated you a better rate with British Gas because we've looked at your contract, we see you coming up, we see you charge your electric car, we think it's going to be sunny for the next 12 months, so we've got you this rate.

unknown: 24:19

Yeah. Yeah.

SPEAKER_00: 24:21

I I think, yeah, I I agree. Those all those things are coming. I think right now, though, for businesses, I you know, the the some of the some of the tasks that we just talked about that you can get agents to do, they they can save development work. So nobody's ever got enough resource to get all the you know, everybody's got technical requirements to be built, or they want their vendor, you know, their supplier of their software to change the the product. Some of these agents sort of allow you to bypass that so you can build effectively your own feature to go to a software tool.

SPEAKER_02: 24:53

So and an individual can do that just for the one task that they're this is where I get dangerous though, because this is where the likes of me, who has a zero technical ability, will produce something that looks great, and then when it breaks, has no idea how to fix it. And that's when I think we rely then on the agent or the AI to do it. If they can't fix it, we're in a world of pain.

SPEAKER_00: 25:14

Well, you say that, but I I one of the you we talked earlier about adoption, I don't think we've completely closed that off. But one of the best ways to get people to adopt their agents is and anybody can try this, uh you can try this at home, as they say, is you if you just give a few line prompts to Chat GPT's agent mode, okay, and you know, set it going, you literally a minute, give it a task, go to these five websites and get me this information, it will it will talk you through what it's doing, it'll say by um it'll say things like I'm clicking on the home button, but I can't find what I'm looking for, so I'm gonna try this, okay? And it could take uh 10-15 minutes to do. Okay. Once you've done it once, you then say to ChatGPT, could you tell me how I could have made it easier for you to have done that and it and it'll say, If you told me that that all the information was on page called this, and give me the URL and the five URLs, I could have done it in a minute. Okay, so they're they are remarkably persistent, so it it it will keep going and and and and you don't get many failures with them. What you do get is you get long, you know, it's like you know, and and yeah, open AI are paying the compute cost for that, but you do get long tasks that take too long.

SPEAKER_02: 26:34

It's a logic tree, I assume. So when it hits a dead end, it'll reverse up and go down the next one until it keeps getting further and further and thinks it's done it.

SPEAKER_00: 26:41

It'll keep trying. So if you give it a really simple prompt, it'll keep going. But if you gave it a more direct prompt, which is go to this page, look at row seven and collect that bit of data, and it's in this format, and then put it into this attached spreadsheet in this format. So clear and clear instructions. Yeah, exactly. Like any any anybody, any assistant or coworker, you give it a clear instruction, it'll just go and be much quicker.

SPEAKER_02: 27:02

So have we got good examples that are out there at the moment of agents that people could copy where they're using it in a specific example? I think there's bits around fraud, is there out there, there's bits around predictive maintenance are here, so it it almost knows when a chiller's gonna break down in a supermarket, so you can replace the motor before it breaks down on the sunny day and you can't sell ice cream. Are there bits like that that people can start to go after?

SPEAKER_00: 27:26

Yeah, well, some of some of those are like forecast engines, so they're like alerts. Yeah. So they're they're giving you an alert. So the so the another way to think of that is you've got tools that are triggering alerts, you know, where they're just you know, they've they're forecasting what's meant to be happened, and if something different happens, they alert you. Yeah. Okay, so the temperature should be 25 degrees and it's it's it's you know, and it's you know, the air con should be 25 degrees and it's it's much colder than that. So then the agent would go and automatically act on that decision, would go and reconfigure the you know, would be a good one.

SPEAKER_02: 28:00

Or call the call the maintenance person to come and fix it.

SPEAKER_00: 28:03

Uh and there there are people using AI to optimise their saving money on their electricity bill by optimising their aircon, so you know, across large estates and large buildings, so that something's monitoring it, it triggers an alert, and instead of it sending a message to the the maintenance person to go up to the fourth floor and turn the the air con down, it just the agent does it. Yeah. So that so the the the agent is is taking a tu is taking an action.

SPEAKER_02: 28:33

Okay, and that uh so yeah, the examples you you talked about, yeah, chatbots, where they're just responding to customer queries and they've come on a bit because they used to be really binary, didn't they, of if you didn't give it exactly what it wanted, it'd just kind of say, Oh, I'll put you through to an agent, or it'd go back into a loop.

SPEAKER_00: 28:53

Uh and and here's the thing. So if I'm talking to a chatbot, and James was alluding to this, and I say to it, you know, can you can you give me effectively FAQs, can you tell me what time you open? Okay, that that to me is not an agent, that's just uh you know, like a smart communication large language model. If you said, Yeah, listen, I've got a problem in my account, you know, I'm frozen out, can you change the passwords, or or something like that? It might ask a few questions to do security.

SPEAKER_02: 29:23

Yeah.

SPEAKER_00: 29:24

Okay. It might not be happy with one of the answers, so it asks the second level question, and then it might come back and say, Yeah, fine, you've you've passed security now. I'm going to reset your password. Can you confirm when you you we've sent you a text, you know, all that sort of stuff. Yeah. It could send the text, it could wait that, and when it's got all the answers, it it's happy, it can reset your password and then send you the whatever, the the the the temporary password. So that's the difference between it that's they're they're good use cases for agents. But the agent has to take some action. It can't just be giving you an answer. That's it.

SPEAKER_02: 29:57

And there's there's lots of insecurity as well, certainly in retail, isn't there? If you shop at a self-checkout now, there's probably a camera above you or on it that's Maybe not turn on for facial recognition, but we'll be in the future to look at age profile. There's some stuff in the news about people being on AI recognising serial uh shoplifters and putting them on banning lists or uh alerting other people on that retail park or in that shopping centre. So all that stuff seems to be, because of the cost challenge, really pushing the user cases of it and the technology. You got any other examples, James, before we close?

SPEAKER_01: 30:31

Yeah, I think I read a case study from Domino's in Australia, which is interesting because the Australian market is generally considered to be one of those that is lagging in AI adoption, like um like the UK. But it was in Domino's and they'd used a camera to monitor pizza quality as it was coming out of the oven. And, you know, a simple automated tool would just tell you the pizzas are coming out bad, go and investigate it. But if you build into that pizzas are coming out a little bit burnt, I'm going to turn the oven down automatically, then it it works more effectively. And there was it was a franchise in Melbourne had used a combination of demand forecasting using AI and some cameras to monitor product quality. They were getting pizza delivery down under six minutes from clicking order on the website through to delivery to house.

SPEAKER_02: 31:17

Better quality, I assume as well.

SPEAKER_01: 31:18

Better quality, quicker, better customer satisfaction, actually enabling them. You know, you almost get the pizza before you've clicked it. It's you know, it's enabling them to that's minority report now. Predicting pizzas, we're all going to start to get Domino's delivery before we order it on the TV. We know we're going to get these orders at at this time. We pre-make them, they're ready to go fresh and hot and the highest quality. You know, it's redefining what you can expect from pizza delivery because it's taking 25 minutes off the delivery time. It's you know, it's really remarkable. And that's where a combination of agents can really enable a business, a physical business selling physical products to make a huge difference to what it does.

SPEAKER_02: 31:56

Perfect. So we will close podcast two on that point. And everybody listening and watching will now be having pizzas delivered that they didn't know they ordered. So enjoy your pizzas. I'm sure it'll be well made. Hopefully, it's not come from Australia because it might be a bit cold. Thank you, Ian. Thank you, James, and I will speak to you again, James, on podcast three.

SPEAKER_00: 32:15

Yes, thank you. Thank you, Sharon.