00:00:01Exploring AI Opportunities in Tech
00:12:59Data Modeling and Application Development
00:27:03Challenges of AI Implementation in Business
00:39:13Innovative Strategies for Software Growth
ABOUT THIS EPISODE
FULL SHOW NOTES
https://www.microsoftinnovationpodcast.com/591
Can AI really transform the way businesses and governments operate? Find out as Ana Welch, Andrew Welch, Chris Huntingford and William Dorrington dive into the latest trends and initiatives driving AI adoption, particularly in the wake of Microsoft's fiscal year priorities and the post-election AI surge in the UK government. Listen to tech guru Benedict Evans dissect the hype versus the real-world application of AI, and get inspired by Sainsbury's innovative use of AI for weather-based inventory management. This episode is brimming with valuable insights that will help you understand the practical implications and benefits of integrating AI into your organization.
But that's not all—we tackle the intricacies of data modeling and application development on platforms like Dynamics and Power Platform. Discover why solid data models are crucial and the common pitfalls citizen developers encounter, such as over-reliance on text fields for key data points. We stress the importance of professional software development practices, including application lifecycle management, testing, and security. Moreover, we highlight the critical role partners play in guiding organizations through these complexities, ensuring effective governance and productivity.
Finally, we address the multifaceted challenges and immense opportunities that come with AI implementation in business. From substantial returns on investment to the essential need for upskilling staff, this episode covers it all. We also examine concerns regarding the accuracy of AI usage metrics and the phenomenon of "AI washing." Concluding with innovative strategies for software growth, we encourage you to keep pushing the boundaries and creating value in your tech endeavors. This episode promises to keep you at the cutting edge of AI and software development, packed with actionable insights and forward-thinking strategies.
If you want to get in touch with me, you can message me here on Linkedin.
Thanks for listening 🚀 - Mark Smith
English
United States
IN THIS EPISODE
SHOW NOTES 🔗
TRANSCRIPT 🔗
WEBVTT
00:00:01.080 --> 00:00:02.685
Welcome to the Ecosystem Show.
00:00:02.685 --> 00:00:05.160
We're thrilled to have you with us here.
00:00:05.160 --> 00:00:11.153
We challenge traditional mindsets and explore innovative approaches to maximizing the value of your software estate.
00:00:11.153 --> 00:00:13.523
We don't expect you to agree with everything.
00:00:13.523 --> 00:00:17.472
Challenge us, share your thoughts and let's grow together.
00:00:17.472 --> 00:00:19.525
Now let's dive in.
00:00:19.525 --> 00:00:20.548
It's showtime.
00:00:20.548 --> 00:00:22.524
Hey, welcome everybody.
00:00:23.167 --> 00:00:27.710
Hello, hello, all five of us are back today, except for the.
00:00:27.710 --> 00:00:28.702
The thing is we.
00:00:28.702 --> 00:00:32.996
We decided we, we don't need will on the show, so we kicked him out.
00:00:32.996 --> 00:00:34.762
We said you will, we go have some sleep.
00:00:34.762 --> 00:00:37.107
And then andrew was complaining.
00:00:37.107 --> 00:00:41.865
So he said andrew, you're out, like we don't need you, and um and chris.
00:00:41.865 --> 00:00:48.634
Um, chris is out getting another tattoo, so that just left Anna and I.
00:00:48.634 --> 00:01:00.027
So you're lucky to have exclusively Anna in the building to hear all about what's top of mind for her at the moment in tech and technology.
00:01:00.027 --> 00:01:02.106
We're coming off an interesting week.
00:01:02.106 --> 00:01:10.191
Fy25 has kicked off, microsoft have just done MCAPs, and so they are setting the priorities for FY25.
00:01:10.191 --> 00:01:19.275
And, lo and behold, ai is kind of featured quite prominently in the FY25 game plan.
00:01:19.275 --> 00:01:23.171
So we might touch on some of those things as we go through today.
00:01:23.171 --> 00:01:26.801
But, anna, what's been going on with you, hey?
00:01:26.900 --> 00:01:40.144
so everything's been really busy lately, maybe also because I work with central government institutions and, as you know, the UK had an election last week.
00:01:40.144 --> 00:01:41.906
Had an election last week.
00:01:42.628 --> 00:01:42.908
Yeah.
00:01:43.287 --> 00:02:07.128
So there was a huge shift that, frankly, I think everybody was able to foresee, but it does mean that a lot of these organizations had things in progress or things that were top of mind for them and they had to pause everything and now they're ramping up again rather rapidly.
00:02:07.128 --> 00:02:13.151
So it's been a little bit chaotic, not to mention exactly what you said.
00:02:13.151 --> 00:02:36.912
You know microsoft's new financial year, new priorities, and everybody, even the strictest, uh central government organization, is talking about AI and they really, really want to at least try it out, get into it, just improve that operational efficiency right now.
00:02:36.912 --> 00:02:44.293
They were under pressure before, but they continue to be under pressure right now and, yeah, they need to do something about it.
00:02:44.855 --> 00:02:45.560
That's interesting.
00:02:45.560 --> 00:02:55.652
It's interesting that I, you know, every company, you know, as you say, it should be on their lips talking about what is the game plan in the era of ai.
00:02:55.673 --> 00:03:03.596
I just want to share something here on screen this is an email, a newsletter, that is put out by benedict evans, and I don't know if you've heard of benedict evans, but this guy's been in.
00:03:03.596 --> 00:03:07.889
I don't know if you've heard of Benedict Evans, but this guy's been in tech a long time.
00:03:07.889 --> 00:03:11.907
If you take a look at the About section on his website you can see 20 years in the space.
00:03:11.907 --> 00:03:37.247
He has 175,000 subscribers on his newsletter and he talks to companies, lectures at companies like Alphabet, amazon, hitachi, all the big players kind of thing out in the market, and so and he, he does this deep analysis of of what's going on in the world and he talks about in this most recent newsletter, the AI summer and the whole.
00:03:37.909 --> 00:03:44.347
You know everybody there's been hundreds of millions of people have tried jet beat chat, gpt, but most of them haven't been back.
00:03:44.347 --> 00:03:45.862
And I feel it Sometimes we live in people have tried chat beat chat, gpt, but most of them haven't been back.
00:03:45.862 --> 00:03:53.463
And I feel that sometimes we live in this bubble of everything's around ai at the moment and it's definitely, you know, taking the oxygen in what we're doing.
00:03:53.463 --> 00:03:56.911
But I'm wondering, is that the case out in?
00:03:56.911 --> 00:04:01.786
You know, general pop and and is the general population really?
00:04:01.786 --> 00:04:05.031
Is ai on their radar as much as it on ours?
00:04:05.031 --> 00:04:10.367
Or is it because we're in this space that we, we like, feel like everybody's on board?
00:04:10.367 --> 00:04:19.533
Like you know, over the last two months my subscriptions to AI models on AI tools have over doubled.
00:04:19.533 --> 00:04:21.865
Like I'm, I'm paying for.
00:04:21.865 --> 00:04:24.665
You know I've had a GPT from day one paid.
00:04:24.665 --> 00:04:25.088
I have a perplexity.
00:04:25.088 --> 00:04:25.572
I paying for.
00:04:25.572 --> 00:04:26.658
You know I've had GPT from day one paid.
00:04:26.658 --> 00:04:29.206
I have perplexity.
00:04:29.206 --> 00:04:29.930
I pay for.
00:04:29.930 --> 00:04:40.142
I have stable diffusion, I pay for on the image side and I have mid-journey, mid-journey, mid-journey, mid-journey.
00:04:40.142 --> 00:04:41.226
Oh for ARCH.
00:04:41.226 --> 00:04:46.110
Right, I feel more engaged than ever in these technologies.
00:04:46.259 --> 00:05:00.709
But going back to what Benedict is saying, there, a lot of companies he's referenced there have done a POC and it died to death, but it hasn't progressed beyond that.
00:05:00.709 --> 00:05:04.607
And then you get these customers that go, we need AI and you're like, oh, what for?
00:05:04.607 --> 00:05:04.971
I don't know that.
00:05:04.971 --> 00:05:05.254
Go, we need ai.
00:05:05.254 --> 00:05:05.819
And you're like, oh, what for?
00:05:05.819 --> 00:05:09.512
I don't know, we just need ai, you know, and it's kind of like.
00:05:09.512 --> 00:05:14.286
I think it's a massive opportunity for consultants, by the way, unbelievable.
00:05:14.286 --> 00:05:17.252
But a consultant has to get really crisp on one.
00:05:17.252 --> 00:05:23.750
How do they uncover ai problems in an organization?
00:05:23.750 --> 00:05:27.860
And what I mean by that is is areas where ai can solve something.
00:05:28.583 --> 00:05:38.190
I heard this case the other day, which I and I love kind of hearing these ones that are unique from the, the field of sainsbury's right, which is a.
00:05:38.752 --> 00:05:55.091
For those of you not in the uk, sainsbury's is like a, a supermarket where you can go get food and stuff, and they are using ai to decide what meat they buy based on the weather conditions.
00:05:55.091 --> 00:06:23.922
For a sainsbury's store, because you think about it, if it's a heat, if it's hot and you know the uk's had a bit of a heat wave at the moment they know that sausages, burger patties and things like that are going to be in higher demand and and so when you think about this and stocking of a shelf in a supermarket, you can see that ai could be absolutely helpful in that scenario by analyzing where the patterns people are going to want to barbecue people are going to.
00:06:23.922 --> 00:06:26.471
You know they're not going to want, let's say, a stew.
00:06:26.471 --> 00:06:30.468
You know meat for a stew or mince or something like that, because they're going to be out and about.
00:06:30.468 --> 00:06:34.528
So let's stock the shelves with that because that product's going to be moved over.
00:06:34.528 --> 00:06:35.533
What might be.
00:06:35.533 --> 00:06:36.415
This is a time of year.
00:06:36.415 --> 00:06:39.480
We would traditionally put that on a full raw chicken.
00:06:39.680 --> 00:06:41.242
Yeah, right, yeah, exactly.
00:06:41.242 --> 00:06:49.507
That's a really good example and I find it really interesting because we didn't even talk about this before.
00:06:49.507 --> 00:06:55.911
But actually, sainsbury's is a model of data management.
00:06:55.911 --> 00:07:40.420
In 2016, 2017, if I remember correctly they were starting to have big problems with matching prices and, mainly, stocking up the shelves and making sure that their shelves have the best, the absolute best targeted audience, but at the same time, trying to be sensible, so you don't understanding the fact that you can't just put named brands, because that's where the profit is, and the fact that you need to, like, cater for everybody.
00:07:41.180 --> 00:07:54.271
So they did a huge machine learning exercise to understand how to stock their shelves in a way that makes sense, in a way that optimizes their storage units.
00:07:54.271 --> 00:08:07.896
They did a huge supply chain exercise where they would be able to offer prognosis over what they will have to buy, even in their storage areas.
00:08:07.896 --> 00:08:22.872
So what this tells us is that it still goes back down to data, because Sainsbury's is able to stock up their stores based on the weather projections in the UK.
00:08:22.872 --> 00:08:39.951
It does not mean that other supermarkets may be able to do it, because Sainsbury's already has the infrastructure and the platform to actually implementing it, and I think this is what potentially customers do not understand.
00:08:39.951 --> 00:08:47.134
Without that baseline, we can build a POC, that's not a problem.
00:08:49.485 --> 00:08:50.168
And then what happens?
00:08:52.102 --> 00:09:16.479
And with regards to how many people use AI tools, I remember at the beginning of last year where my best friend was very worried about artificial intelligence and how that's going to steal, you know, our jobs, and I was like, well, I'm not sure it's quite that way.
00:09:16.479 --> 00:09:28.347
And, yes, maybe some professions will be changing or disappearing completely, but others, you know, potentially will be enriched, whatever.
00:09:28.347 --> 00:09:37.067
I didn't have enough information to make a very good point either, especially because fear is a very, very powerful thing, right.
00:09:37.208 --> 00:09:37.811
Yeah, yeah.
00:09:38.659 --> 00:09:42.527
Well, fast forward to this year she has been.
00:09:42.527 --> 00:09:42.606
She.
00:09:42.606 --> 00:09:56.934
She's very smart individual, so she has been promoted twice and now she's in a management position and her number one job is to liaise with people.
00:09:56.934 --> 00:09:58.280
And what does that mean?
00:09:58.280 --> 00:09:59.780
Writing emails.
00:09:59.780 --> 00:10:04.302
And I'm like why don't you use AI to draft those damn emails?
00:10:04.302 --> 00:10:09.865
Because otherwise you're going to be working 14 hour days.
00:10:09.865 --> 00:10:12.527
You're never going to finish if you don't get some help.
00:10:12.527 --> 00:10:29.216
This is where I feel people in organizations can use AI, if they realize what's the real use case, not what they want to play with what is the real use case in their day-to-day lives.
00:10:29.775 --> 00:10:43.530
It's an interesting one, because AI you know a lot of what has been talked about is not a single thing, and one of the things that Benedict says in his article is it a product?
00:10:43.530 --> 00:10:45.485
What is it actually?
00:10:45.485 --> 00:10:53.966
And and because I've been using this metaphor that I feel like it is like electricity, and so therefore, it's an enabler of everything.
00:10:53.966 --> 00:10:56.561
It's an enabler of my toast that I eat for breakfast.
00:10:56.561 --> 00:11:00.490
It's an enabler of the you know, the fried bacon that I have.
00:11:00.490 --> 00:11:02.822
It's an enabler of the heat that I have in my house.
00:11:02.822 --> 00:11:04.283
It's it's an enabler of the heat that I have in my house.
00:11:04.283 --> 00:11:09.812
It's an enabler of so many things lighting, et cetera and so one, how do you ROI that?
00:11:09.912 --> 00:11:55.248
And I know Chris Huntingford, if he was with us, would have a lot to say on ROI and AI, and so I feel it is really a different paradigm, it's a different mindset and, like we've talked about the power platform, one of the things that have come up recently again for me is the, the difference between basic licensing and premium licensing, and this concept of of uh software economics, which is when you build a solution, and I was just yesterday talking to a company that has, I think it was a either 11 or 12,000 employees, 6,000 offices around the world and they are using the Power Platform on premium.
00:11:55.828 --> 00:12:07.634
You know for that organization and you know this is a corporate end customer using this technology, um, uh, out of Portugal.
00:12:07.634 --> 00:12:18.302
And what was incredible is that they said that the thing is, is that when we we did the ROI on the first project, right, it was a no brainer.
00:12:18.302 --> 00:12:23.347
But they said now every app we build from this point on, we're not having to think of who do we need to license?
00:12:23.347 --> 00:12:24.152
They're licensed already, right, the.
00:12:24.152 --> 00:12:24.895
Do we need to license?
00:12:24.895 --> 00:12:26.100
They're licensed already.
00:12:26.100 --> 00:12:27.563
Right, the right, the.
00:12:27.563 --> 00:12:28.525
Everybody's licensed.
00:12:28.546 --> 00:12:59.471
So now they're like well, the more solutions that we build that we architect now on this platform, we have diminishing costs as such, because it gets extrapolated right out across across that you build the economy at scale, and it's not just the amount of applications that you have to build, or flows or connectors or pages or whatever you're building, but it's also the supporting framework of it all.
00:12:59.471 --> 00:13:05.708
Right, you set up Android once you will always have to log into your applications.
00:13:05.708 --> 00:13:12.071
Right Done, you have put in that information.
00:13:12.071 --> 00:13:17.044
Once you will always be able to intelligently search for it.
00:13:17.044 --> 00:13:27.054
No matter what application that data sits in, no matter who touches it, you will be able to extract it back out.
00:13:27.054 --> 00:13:41.212
Our friend Chris said in his presentation in the latest presentation that I've seen of him, he was like you know how you like, type in data or import data into a system.
00:13:41.212 --> 00:13:45.331
The really cool thing would be to be able to find it again.
00:13:45.331 --> 00:13:47.413
How true is that?
00:13:47.774 --> 00:14:26.859
Yeah, yeah yeah, like that's so funny, but so true, right, and all of these things are things that you have probably built with your project managers and, I don't know, your BAs, your UX team, who know what to do always.
00:14:26.859 --> 00:14:27.666
It's.
00:14:27.666 --> 00:14:31.174
The scale is huge.
00:14:31.174 --> 00:14:32.157
It's humongous.
00:14:32.764 --> 00:14:59.379
What are you doing in the space of customers that have perhaps really leaned into the concept of personal productivity with a power platform and realizing that there's limits to how far you can go there and that a non-trained software developer builds software quite differently than a trained software developer?
00:14:59.379 --> 00:15:29.004
Right, and you know, standard rigor that we have around things like application lifecycle management, around testing in all its different formats, around documentation, supportability, all those type of robust things anything down to security-based testing of an application is what you have in a traditional software world.
00:15:29.004 --> 00:15:34.134
And then you've got the personal productivity and, dare I say it, citizen dev comes to mind.
00:15:34.134 --> 00:15:38.482
There's limits to how far one can go with that.
00:15:38.482 --> 00:15:40.067
What are you seeing around that?
00:15:40.128 --> 00:16:19.254
and moving into the area of more, um, you know, enterprise, like going up that ladder to into the enterprise more you know, in the last few months I've seen it a lot from from organizations who are starting, who have started rather straight on a mature platform, either on Dynamics or Power Platform, but with data first, to customers who have not, and they've leveraged SharePoint lists and Excel spreadsheets in order to create their productivity apps.
00:16:19.254 --> 00:16:25.352
They're all trying to navigate this world and, I'm not going to lie, it is not easy.
00:16:25.352 --> 00:16:27.493
It's like actually very complicated.
00:16:27.493 --> 00:16:40.078
From the organization who swallowed the citizen developer pill and they thought that actually we're just going to do all of this, all of this, ourselves.
00:16:40.078 --> 00:16:46.197
We only need a partner to support the stuff that we're doing.
00:16:46.304 --> 00:16:50.155
I know that you've said in previous iterations that the partner is dead.
00:16:50.785 --> 00:17:02.640
I would say that not yet, because, listen, these organizations they're not prepared for the level of assurance and diligence that the projects need.
00:17:03.384 --> 00:17:35.244
They're not ready to write, but they need to put in a lot of time to learn basic things like how to write a user story or how to ensure that it will travel correctly through all of your environments, how to make sure that you're not, you know, sending random emails to a production database.
00:17:35.244 --> 00:18:00.173
So that's like a huge, teething pain where I believe partners need to be consultants.
00:18:00.173 --> 00:18:18.012
You know Power Platform icon and started building applications and they're all hanging on SharePoint lists and Excel spreadsheets and some of them are like, oh, actually wait, we have a few licenses for you know premium Power Platform, but what do we do next?
00:18:18.012 --> 00:18:19.411
Who do we give them to?
00:18:19.411 --> 00:18:20.911
Do we buy more?
00:18:20.911 --> 00:18:35.845
They are truly recognizing the fact that they need the functionality, that they need the power, the discipline, the government, the governance, the security underneath it all, but it's very, very hard to get started.
00:18:35.845 --> 00:18:43.539
So that was a really, really, really long way of saying it's complicated to get started in these situations.
00:18:43.539 --> 00:18:45.645
Where do you think one should get started?
00:18:45.826 --> 00:19:03.119
actually, from your experience, yeah, well, when I ever think about where do you get started and it's not explicitly about what you're talking about I always think at the very beginning it's a very good place to start.
00:19:03.119 --> 00:19:08.888
When you sing, you begin with ABC, sorry, wrong show, this is a.
00:19:08.888 --> 00:19:10.294
It's not that story.
00:19:10.294 --> 00:19:12.482
No, I always think about data.
00:19:12.482 --> 00:19:13.246
Right, as in.
00:19:13.246 --> 00:19:18.487
I was brought up in that you model data up to the app layer, not the other way around.
00:19:18.487 --> 00:19:43.287
And I feel what my observation and I've worked with a few architects recently I am concerned about, about how even architects looking at building a solution from a data model perspective and I'm talking about going down to a field level, or is it a column level now, sorry, I still use the old nomenclature, um, because I was brought up with the old nomenclature.
00:19:43.287 --> 00:20:03.592
So the text field I hate it with a vengeance, right, yeah, and and the reason sorry, I don't hate it, like it's obviously has its place, but I hate it when it's used as a catch-all, right, and I see this, like you know, back back in in the early days of Dynamics CRM.
00:20:03.592 --> 00:20:18.232
There was a field in there, I just from memory and it said something like, let's say it was on a contact record and you said this contact was a manager and there was a field that said what's their executive assistant or no, it was their spouse, who was their spouse?
00:20:18.232 --> 00:20:19.615
And it was a text field.
00:20:19.615 --> 00:20:24.913
And I'm like, if it's another person, why isn't that a look up to a contact entity?
00:20:24.913 --> 00:20:28.123
Because now you leave intelligence in the system.
00:20:28.123 --> 00:20:32.596
Yeah, right, because you know a contact is more than a name.
00:20:32.596 --> 00:20:36.106
Right, it has other things, like an email address, a phone number.
00:20:36.106 --> 00:20:37.288
Why do I just want to know their name?
00:20:37.288 --> 00:20:39.733
Was it just so I drop it into a conversation?
00:20:39.733 --> 00:20:40.115
Anyhow?
00:20:40.115 --> 00:20:45.939
That aside, I see it on things like informed data collection online or anywhere where they go.
00:20:45.939 --> 00:21:13.098
The country field is a text field and you're like, imagine somebody reporting on that field now and all the spelling mistakes, um, that will come just with people typing the country name incorrectly, as opposed to that being you know a pre-populated list of what it could be, and then, of course, with longer lists, how do you make the interface so that you don't feel like you're scrolling through a massive drop-down list?
00:21:13.098 --> 00:21:20.191
But I'm like I feel that there needs to be more rigor around choosing the right type of field.
00:21:20.191 --> 00:21:21.955
Is it field or column?
00:21:21.955 --> 00:21:24.763
What am I meant to be saying I think it's column.
00:21:24.763 --> 00:21:26.028
Yeah, what type of column?
00:21:26.028 --> 00:21:26.189
What?
00:21:26.189 --> 00:21:34.837
What is the column type not name, but type that I should be selecting to make my system stay intelligent beyond just me building the application.
00:21:34.837 --> 00:21:36.221
Now, what is it going to mean from a reporting perspective?
00:21:36.221 --> 00:21:38.327
Because is it going to mean from a reporting perspective?
00:21:38.327 --> 00:21:45.288
Because it's going to wrap metadata around it based on, you know, the field, type, um, the, the column type you specify.
00:21:45.288 --> 00:21:47.894
Yeah, I just notice that there's.
00:21:49.516 --> 00:22:01.912
There's often not that level of rigor, yes to, and I always say the worst critic you can have is your customer, because your customer doesn't know what good looks like.
00:22:01.912 --> 00:22:05.826
Right, you can show them something and they might think, wow, that looks good.
00:22:05.826 --> 00:22:19.508
I always say the best critic is my competitor, yeah, somebody with the same skills as me in a different organization, because they're going to apply the rigor at the same knowledge level as I have to go.
00:22:19.508 --> 00:22:20.951
Is that best practice?
00:22:20.951 --> 00:22:23.019
Is that the best way of doing this?
00:22:23.019 --> 00:22:27.048
Is that going to make the system more intelligent or less intelligent, etc.
00:22:27.048 --> 00:22:27.450
Over time?
00:22:28.050 --> 00:22:32.666
And so I feel that there needs to be a robustness around that area.
00:22:32.666 --> 00:22:35.550
So when you say where do I start, I'm always thinking about you know.
00:22:35.550 --> 00:22:39.336
How's this data going to be used outside of just it's?
00:22:39.336 --> 00:22:42.299
Now, you know application, how is it?
00:22:42.299 --> 00:22:56.355
And that's why I love andrew's comment that the the power platform is not an app story, it's a data story, and I think, until people really start to realize that, and then and then my question is how do you learn to get good about data, how do you learn as an?
00:22:56.355 --> 00:23:07.417
What I mean is much more beyond just the power platform, but you know how to catalog it, reference it, how to make it available, how to sanitize it, how to, uh, archive it right.
00:23:07.417 --> 00:23:11.326
A lot of organizations have no archiving strategy, right?
00:23:12.048 --> 00:23:17.016
especially if you're young yeah, like if you started 15 years ago.
00:23:17.016 --> 00:23:22.933
You kind of know about data because you have to yes, there's no way around it.
00:23:22.933 --> 00:23:37.634
But if you didn't, if you're just trying to do something right now, it's really really hard to know about data and, even more tragic, it's really hard to show others the value of data.
00:23:37.634 --> 00:23:46.936
So I agree with you, we do start with data, we start with the data model, we start with you know, controls around it, guardrails.
00:23:46.936 --> 00:23:59.185
We start with the idea that we want to know more than that person's name, that we actually want to know her date of birth to send her a birthday card?
00:23:59.185 --> 00:23:59.946
I don't know.
00:24:00.528 --> 00:24:05.298
Her hair color, I don't care, but we need to know more about that.
00:24:05.298 --> 00:24:06.465
Yeah, so that's the idea.
00:24:06.465 --> 00:24:33.891
However, if I am the executive sponsoring this, in the age of AI, when I can see everything right now and instant gratification is even at higher levels, we need to come up with a way to make data sexy, to make it attractive, to make it the cherry on top, something that people want to see.
00:24:34.731 --> 00:24:43.682
Yeah, I think that's easier than what people realize to make it sexy.
00:24:43.682 --> 00:24:53.855
And the thing is that, you know, I remember on one project where the customer said for every item of data my salespeople put into the system, I want five benefits back.
00:24:53.855 --> 00:24:56.352
Right, and it really made you think, right.
00:24:56.705 --> 00:24:58.999
Yeah, right, benefits back, right so and it really made you think right, yeah, right.
00:24:58.999 --> 00:25:10.277
What you realize pretty quickly is that it's actually achievable, especially when you've got the ability to use a connector and bring in other data so you can enrich that data set by getting a small amount.
00:25:10.277 --> 00:25:21.133
So you know, for example, in the uk if you put somebody's postcode in, it narrows down to pretty much the house on the street even without knowing the street number.
00:25:21.133 --> 00:25:24.335
There's a high degree of accuracy which you can look up and validate.
00:25:24.335 --> 00:25:28.772
Things like phone number format Don't ask me to put in a certain format.
00:25:28.772 --> 00:25:33.997
I'll put it in whatever format that I think of, programmatically updated to the format that you want.
00:25:33.997 --> 00:25:43.758
Right, this stuff's all possible nowadays, but it's often that kind of polish and stuff that we leave off in the speed, I suppose, of getting things done.
00:25:43.758 --> 00:25:46.817
You know, don't make me enter the same piece of data twice.
00:25:46.817 --> 00:25:51.231
If it's this in the organization, bring it in from where it exists.
00:25:51.231 --> 00:25:58.000
Let's not create multiple copies um of or make people fill in data multiple times if it's available.
00:25:58.020 --> 00:26:02.276
Right, show it to me, it's one of my biggest pet peeves is that you know this information about me.
00:26:02.276 --> 00:26:04.231
You know whoever I'm buying from.
00:26:04.231 --> 00:26:06.069
Why are you asking me again?
00:26:06.069 --> 00:26:07.133
You know this?
00:26:07.133 --> 00:26:10.050
Yes, exactly, and that's where the rigor needs to come in.
00:26:10.644 --> 00:26:13.914
You know I just bought paint and that it was white.
00:26:13.914 --> 00:26:16.413
Stop trying to sell me white paint.
00:26:16.884 --> 00:26:18.932
Yeah, yeah, yeah, so true.
00:26:18.932 --> 00:26:35.576
Another thing I just want to share and I'll just bring up my screen here again is this comment, sorry, this slide that I saw this week from Satya, and I'm loving this right because back on our AI again.
00:26:35.576 --> 00:26:38.213
Ai is the greatest commercial opportunity in today's economy.
00:26:38.213 --> 00:26:44.892
This next generation of AI will reshape every software category in every business, including our own.
00:26:44.892 --> 00:26:56.096
From Satya, right, and what I'm seeing is that, and you can see here, bloomberg Intelligence 2023 is where the data on the right is referenced, and then the Microsoft Work Trend Index, 2024.
00:26:56.096 --> 00:27:02.955
But look at these numbers 280 billion an additional software revenue opportunity based on rising demand of ai products.
00:27:02.955 --> 00:27:06.671
71 say their companies use ai now.
00:27:06.671 --> 00:27:08.855
And I bring this up for a couple of reasons.
00:27:08.996 --> 00:27:16.949
One, the opportunity is massive right yeah you know, for every dollar invested, three to eight times the return and that that's really that 92.
00:27:16.949 --> 00:27:27.749
There there is the partner opportunity on this, and so there's a real need for partners to upskill their staff right, because you can't?
00:27:27.749 --> 00:27:31.837
And I feel there's two kind of skilling types that are needed.
00:27:31.837 --> 00:27:55.648
What you see on Microsoft Learn is how to use our technology skilling, and I see that the world and the community that we're in seems to have indexed heavily on that, but there's not a lot on like first principle, thinking around data outside of a piece of software as an example, right or or or any of these type of things.
00:27:55.648 --> 00:28:01.951
Like you know, you can technically get into what purview does, but what's the application?
00:28:01.951 --> 00:28:04.336
What's the broader implication of doing it?
00:28:04.336 --> 00:28:05.709
Why is it important?
00:28:05.709 --> 00:28:08.961
And it's answering all these business value questions.
00:28:08.961 --> 00:28:10.865
I think there's a lot more focus.
00:28:10.865 --> 00:28:16.653
The other thing is is that this number here of 71% say their companies use AI.
00:28:16.653 --> 00:28:33.936
I know, I feel I expect that's kind of bullshit, and why I say that is that the telemetry metrics that people use sometimes inflate these numbers or deceive these numbers, and it's happened for years.
00:28:33.936 --> 00:28:40.035
So, for example, if you took YouTube, tiktokiktok and facebook and you had a video on each.
00:28:40.035 --> 00:28:43.925
They will all count the number of views of that differently.
00:28:43.925 --> 00:28:48.496
So youtube, for example, say if it's watched for 30 seconds, that's a view.
00:28:48.496 --> 00:28:52.874
Tiktok will say well, if it loads on screen, that's a view, doesn't matter if it was played.
00:28:52.874 --> 00:28:54.467
If it loaded on screen, it's a view.
00:28:54.467 --> 00:28:56.410
Facebook might be five seconds.
00:28:56.410 --> 00:29:00.679
They all say this is my view, time right, and it's just sell a story.
00:29:00.798 --> 00:29:10.588
And and when we look at co-pilot, co-pilot, for example, you get the scenario that how do you tell whether you're getting adoption inside an organization?
00:29:10.588 --> 00:29:13.076
Is adoption measuring the right thing?
00:29:13.076 --> 00:29:15.644
So I I saw this example just yesterday.
00:29:15.644 --> 00:29:16.722
Where they're going, oh, look at this, we're measuring the word is the.
00:29:16.722 --> 00:29:16.972
So I saw this example just yesterday.
00:29:16.972 --> 00:29:28.954
Where they're going, oh, look at this, we're measuring Word is the most common copilot use case and I'm like hang on a second, how is it telling that somebody is using it in Word?
00:29:29.778 --> 00:29:32.125
Oh, a plugin fires when Word opens.
00:29:32.125 --> 00:29:33.326
It calls the copilot.
00:29:33.326 --> 00:29:40.412
So actually nobody used AI, just the software called and said you know, is the co-pilot connector alive?
00:29:40.412 --> 00:29:42.112
And we're counting that as use.
00:29:42.112 --> 00:29:50.405
Come on, nobody even used it, it just happened to be a trigger in the software and that's why I'm like these metrics sometimes, and it's.
00:29:50.405 --> 00:30:03.432
You know AI is a fuzzy term because AI has been around since the 60s, the 1960s, right, and it's like now, anything that's kind of an algorithm or a if, then statement, that's ai.
00:30:03.432 --> 00:30:13.767
You know that people are categorizing that as broadly ai as well, because the computer made a decision but it does make sense, like it does remember how we have.
00:30:14.426 --> 00:30:30.423
Um, what's the name of the app Office Vibe, I want to say, I don't know the one that tells you oh, you have collaborated with your peers for three hours a day, or you've had focused work for such and such.
00:30:30.423 --> 00:30:35.464
Well, that can definitely measure whether you're so.
00:30:35.464 --> 00:30:50.021
Are we measuring if someone has Teams Premium or Copilot for Microsoft 365, that they've used Copilot just because they've been in Teams or they did a Bing search or whatever they did?
00:30:50.021 --> 00:30:51.672
Because that wouldn't be accurate.
00:30:52.246 --> 00:30:52.707
Yeah accurate?
00:30:52.727 --> 00:30:54.791
yeah, if you look at it from one perspective.
00:30:54.791 --> 00:31:14.765
But if you look at it from your premises perspective, that says ai will be like electricity we will not feel it, we will not see it, it will just toast our toast brilliant then it's true yeah, valid, valid, you have a good point.
00:31:15.545 --> 00:31:18.747
and, and, and there is that whole thing around digital exhaust, right?
00:31:18.747 --> 00:31:38.096
All the stuff that goes on that you don't think about, which you know the graph is great at understanding and picking up and making sense of the information reported on might be more helpful in other contexts, even though that individual might not, you know, be doing it.
00:31:38.096 --> 00:31:43.900
But then, you know, will AI become so immersive that we're not on purpose using it?
00:31:43.900 --> 00:31:47.922
It's just like I don't think of, when I turn on a light switch that I am going to use electricity.
00:31:48.001 --> 00:31:48.883
No, I just lit the room.
00:31:50.724 --> 00:31:51.045
Yeah so I agree.
00:31:51.045 --> 00:31:56.548
And then shooting out a figure like 71% can make people believe.
00:31:56.548 --> 00:32:14.021
Wait a second, if this is 71% already, you know of organizations using it already why should I even bother to move my data into and model it into like a sensible program of work?
00:32:14.021 --> 00:32:14.982
Why should I do that?
00:32:14.982 --> 00:32:23.957
That looks like several millions of investment and it doesn't make sense because my stakeholders will not see it straight away.
00:32:23.957 --> 00:32:34.568
So maybe I should just ignore it and use AI the way I want to, because everybody is warning me that bad things are going to happen, but I've not seen them yet.
00:32:36.935 --> 00:32:37.136
Yeah.
00:32:38.386 --> 00:32:39.730
And I am worried about that.
00:32:39.730 --> 00:32:55.615
You know, in my day-to-day work I feel like instead of ramping up and investing and grown-up, problem solving organizations are taking corners already.
00:32:56.265 --> 00:33:08.230
Yeah, and I suppose in this probably we're almost, we're out of time, but yes, another thing that I find concerning is ai washing and yeah, what is?
00:33:08.269 --> 00:33:09.211
that tell me.
00:33:09.211 --> 00:33:10.755
So ai washing is this.
00:33:10.755 --> 00:33:24.134
You know, when I was in london, um, there was a, there was a pride event on and something that came out in the media was was, um, gay washing, uh, your business, right.
00:33:24.134 --> 00:33:32.715
And so businesses come out with their stance on, you know, inclusion and things like that and we're kind of riding the bandwagon.
00:33:32.715 --> 00:33:33.877
We've seen it in the.
00:33:33.877 --> 00:33:37.972
So it's kind of like we're seen to be doing the right thing, whether they're doing it or not.
00:33:37.972 --> 00:33:49.249
And we see the same thing in sustainability, right, in that whole scenario where we're ticking the boxes for the marketing, um, and then you see what's going on beside, behind the scenes.
00:33:49.249 --> 00:33:53.357
I've seen it in the whole area of women in the workplace.
00:33:53.357 --> 00:34:10.791
You know, oh, you know, we so appreciate equal opportunity and everything, and then women go on maternity leave and their job's gone when they come back, right, it's like the chairs got shifted and it's kind of like to the public, it's all, yes, we're this.
00:34:11.246 --> 00:34:14.293
Behind the scenes, it's like still cutthroat as ever.
00:34:14.293 --> 00:34:17.472
There's no adhesion to it.
00:34:17.472 --> 00:34:25.396
So when I look at it in the terms of AI, I see these companies now and all their marketing is AI, this, co-pilot, that AI, this.
00:34:25.396 --> 00:34:28.592
And I look at the company and go okay, how long have you been in business?
00:34:28.592 --> 00:34:37.717
15 years, oh so, and email's been one of the ones for me, right, like how do I get a real optimized email experience using AI?
00:34:37.717 --> 00:34:54.514
And I look at these companies that have got these AI tools and you're like no, you've been doing email classification for 15 years and all of a sudden, because AI has become the trendy word, you're now AI infused and it's kind of like it's like you know the whole.
00:34:54.514 --> 00:34:58.233
Responding to an RFX, it's like yes, we've got that feature.
00:34:58.233 --> 00:35:04.592
And it's like the customer's expecting a feature this big and you're meaning it's this big.
00:35:04.965 --> 00:35:26.500
But it ticks the box of saying, yes, it has AI in it, and that's why I think there's so much AI washing going on in the market, where people are calling their products, they're doing startups that are, they're riding the ai bandwagon, but there's no, you know, there's no substance to what's behind it.
00:35:26.500 --> 00:35:31.652
They're just using it as a marketing phrase rather than it's intelligent yeah, I see that.
00:35:31.771 --> 00:35:33.193
I see that a lot as well.
00:35:33.193 --> 00:35:39.280
I'm not sure I see it a lot in my customers, in our customers.
00:35:39.280 --> 00:36:01.222
I find my customers to be incredibly honest and very much set on good intentions Like I've never seen organizations really, really, really want to work for the greater good.
00:36:01.222 --> 00:36:08.338
However, I'm seeing a lot of partners go on and on and on with AI and what they do.
00:36:09.079 --> 00:36:09.260
Yeah.
00:36:09.605 --> 00:36:16.875
You know, with these tools for their customers, and I am wondering where are these customers who are doing extra?
00:36:16.875 --> 00:36:23.034
You know extra work in AI and it's also intelligent.
00:36:23.034 --> 00:36:25.478
Why aren't they in the news?
00:36:25.478 --> 00:36:27.931
And in fact, I did try to study this.
00:36:27.931 --> 00:36:46.739
I've listened to podcasts and I've read magazines and publications and some more obscure than others and the reality is nobody's offering any sort of consistent result that can hold water.
00:36:46.739 --> 00:37:01.376
You know that you can say yeah, yeah, okay, so I can see that organizations really are using AI, rather than 71% of organizations may have co-pilot in the corner of their being.
00:37:02.137 --> 00:37:03.681
Yeah, yeah, yeah.
00:37:03.681 --> 00:37:06.273
Well, that's all we've got time for today.
00:37:06.273 --> 00:37:11.056
Has anyone picked up the paid or the unpaid placement through the show?
00:37:11.056 --> 00:37:14.114
Can I have a look at your hands, anna?
00:37:14.114 --> 00:37:16.451
Are you also wearing a ring?
00:37:16.632 --> 00:37:17.996
that looks very similar to my ring.
00:37:18.786 --> 00:37:20.713
Oh my gosh, look at that.
00:37:22.085 --> 00:37:22.586
What's what?
00:37:22.648 --> 00:37:24.434
is what is this ring that we're wearing?
00:37:24.434 --> 00:37:26.541
Anna, it is, you sold me on it.
00:37:27.164 --> 00:37:30.030
Indeed, and we are like you know.
00:37:30.030 --> 00:37:38.371
My relationship with the aura ring um is a longer living than my relationship with my husband.
00:37:38.371 --> 00:37:47.159
Wow, I've had this ring since the very first generation of smart rings like that.
00:37:47.159 --> 00:37:52.074
It's got like little sensors that can tell me all sorts of things about myself.
00:37:52.074 --> 00:38:00.965
From how will I sleep about myself?
00:38:00.965 --> 00:38:03.672
From how will I sleep on whether I should do like some housework or go for a walk or do whatever, and it's like super inclusive.
00:38:03.672 --> 00:38:08.972
It doesn't want me to go on a spin class or something, but it does want me to move around.
00:38:08.972 --> 00:38:10.615
It tells me if I'm stressed.
00:38:10.615 --> 00:38:21.376
I've got like meditation techniques on my phone if I want to use them, but if I don't want to use them, it's smart enough that it doesn't push them on me.
00:38:21.376 --> 00:38:27.398
It's sort of like a smartwatch, but 10 times better, in my opinion.
00:38:27.398 --> 00:38:29.913
It's awesome, it's fabulous.
00:38:30.666 --> 00:38:46.590
I agree, I've had it now since Dynamics Mines or thereabouts, when Anna introduced it to me, and it's permanently on me and that's what I like about it over, you know, because I have an Apple Watch as well, but it needs to be charged so often and so it's not on me.
00:38:46.590 --> 00:38:55.032
But the beauty of the ring is that the only time it goes on the charger is when I'm in the shower, and that's enough, and it's like got about seven days charged.
00:38:55.032 --> 00:39:04.775
Now, this is not a paid placement, this is just us geeking out over our wearable technology.
00:39:04.775 --> 00:39:07.081
Um, and with that, anna, it's been a pleasure to just have us talking on the show.
00:39:07.101 --> 00:39:07.824
We don't need those other guys um.
00:39:07.844 --> 00:39:12.170
It was really lovely ciao, ciao, ciao, bye, bye.
00:39:12.170 --> 00:39:14.894
Thanks for tuningcosystem Show.
00:39:14.894 --> 00:39:20.931
We hope you found today's discussion insightful and thought-provoking, and maybe you had a laugh or two.
00:39:20.931 --> 00:39:26.952
Remember your feedback and challenges help us all grow, so don't hesitate to share your perspective.
00:39:26.952 --> 00:39:33.025
Stay connected with us for more innovative ideas and strategies to enhance your software estate.
00:39:33.025 --> 00:39:37.096
Until next time, keep pushing the boundaries and creating value.
00:39:37.096 --> 00:39:38.670
See you on the next episode.
00:00:01.080 --> 00:00:02.685
Welcome to the Ecosystem Show.
00:00:02.685 --> 00:00:05.160
We're thrilled to have you with us here.
00:00:05.160 --> 00:00:11.153
We challenge traditional mindsets and explore innovative approaches to maximizing the value of your software estate.
00:00:11.153 --> 00:00:13.523
We don't expect you to agree with everything.
00:00:13.523 --> 00:00:17.472
Challenge us, share your thoughts and let's grow together.
00:00:17.472 --> 00:00:19.525
Now let's dive in.
00:00:19.525 --> 00:00:20.548
It's showtime.
00:00:20.548 --> 00:00:22.524
Hey, welcome everybody.
00:00:23.167 --> 00:00:27.710
Hello, hello, all five of us are back today, except for the.
00:00:27.710 --> 00:00:28.702
The thing is we.
00:00:28.702 --> 00:00:32.996
We decided we, we don't need will on the show, so we kicked him out.
00:00:32.996 --> 00:00:34.762
We said you will, we go have some sleep.
00:00:34.762 --> 00:00:37.107
And then andrew was complaining.
00:00:37.107 --> 00:00:41.865
So he said andrew, you're out, like we don't need you, and um and chris.
00:00:41.865 --> 00:00:48.634
Um, chris is out getting another tattoo, so that just left Anna and I.
00:00:48.634 --> 00:01:00.027
So you're lucky to have exclusively Anna in the building to hear all about what's top of mind for her at the moment in tech and technology.
00:01:00.027 --> 00:01:02.106
We're coming off an interesting week.
00:01:02.106 --> 00:01:10.191
Fy25 has kicked off, microsoft have just done MCAPs, and so they are setting the priorities for FY25.
00:01:10.191 --> 00:01:19.275
And, lo and behold, ai is kind of featured quite prominently in the FY25 game plan.
00:01:19.275 --> 00:01:23.171
So we might touch on some of those things as we go through today.
00:01:23.171 --> 00:01:26.801
But, anna, what's been going on with you, hey?
00:01:26.900 --> 00:01:40.144
so everything's been really busy lately, maybe also because I work with central government institutions and, as you know, the UK had an election last week.
00:01:40.144 --> 00:01:41.906
Had an election last week.
00:01:42.628 --> 00:01:42.908
Yeah.
00:01:43.287 --> 00:02:07.128
So there was a huge shift that, frankly, I think everybody was able to foresee, but it does mean that a lot of these organizations had things in progress or things that were top of mind for them and they had to pause everything and now they're ramping up again rather rapidly.
00:02:07.128 --> 00:02:13.151
So it's been a little bit chaotic, not to mention exactly what you said.
00:02:13.151 --> 00:02:36.912
You know microsoft's new financial year, new priorities, and everybody, even the strictest, uh central government organization, is talking about AI and they really, really want to at least try it out, get into it, just improve that operational efficiency right now.
00:02:36.912 --> 00:02:44.293
They were under pressure before, but they continue to be under pressure right now and, yeah, they need to do something about it.
00:02:44.855 --> 00:02:45.560
That's interesting.
00:02:45.560 --> 00:02:55.652
It's interesting that I, you know, every company, you know, as you say, it should be on their lips talking about what is the game plan in the era of ai.
00:02:55.673 --> 00:03:03.596
I just want to share something here on screen this is an email, a newsletter, that is put out by benedict evans, and I don't know if you've heard of benedict evans, but this guy's been in.
00:03:03.596 --> 00:03:07.889
I don't know if you've heard of Benedict Evans, but this guy's been in tech a long time.
00:03:07.889 --> 00:03:11.907
If you take a look at the About section on his website you can see 20 years in the space.
00:03:11.907 --> 00:03:37.247
He has 175,000 subscribers on his newsletter and he talks to companies, lectures at companies like Alphabet, amazon, hitachi, all the big players kind of thing out in the market, and so and he, he does this deep analysis of of what's going on in the world and he talks about in this most recent newsletter, the AI summer and the whole.
00:03:37.909 --> 00:03:44.347
You know everybody there's been hundreds of millions of people have tried jet beat chat, gpt, but most of them haven't been back.
00:03:44.347 --> 00:03:45.862
And I feel it Sometimes we live in people have tried chat beat chat, gpt, but most of them haven't been back.
00:03:45.862 --> 00:03:53.463
And I feel that sometimes we live in this bubble of everything's around ai at the moment and it's definitely, you know, taking the oxygen in what we're doing.
00:03:53.463 --> 00:03:56.911
But I'm wondering, is that the case out in?
00:03:56.911 --> 00:04:01.786
You know, general pop and and is the general population really?
00:04:01.786 --> 00:04:05.031
Is ai on their radar as much as it on ours?
00:04:05.031 --> 00:04:10.367
Or is it because we're in this space that we, we like, feel like everybody's on board?
00:04:10.367 --> 00:04:19.533
Like you know, over the last two months my subscriptions to AI models on AI tools have over doubled.
00:04:19.533 --> 00:04:21.865
Like I'm, I'm paying for.
00:04:21.865 --> 00:04:24.665
You know I've had a GPT from day one paid.
00:04:24.665 --> 00:04:25.088
I have a perplexity.
00:04:25.088 --> 00:04:25.572
I paying for.
00:04:25.572 --> 00:04:26.658
You know I've had GPT from day one paid.
00:04:26.658 --> 00:04:29.206
I have perplexity.
00:04:29.206 --> 00:04:29.930
I pay for.
00:04:29.930 --> 00:04:40.142
I have stable diffusion, I pay for on the image side and I have mid-journey, mid-journey, mid-journey, mid-journey.
00:04:40.142 --> 00:04:41.226
Oh for ARCH.
00:04:41.226 --> 00:04:46.110
Right, I feel more engaged than ever in these technologies.
00:04:46.259 --> 00:05:00.709
But going back to what Benedict is saying, there, a lot of companies he's referenced there have done a POC and it died to death, but it hasn't progressed beyond that.
00:05:00.709 --> 00:05:04.607
And then you get these customers that go, we need AI and you're like, oh, what for?
00:05:04.607 --> 00:05:04.971
I don't know that.
00:05:04.971 --> 00:05:05.254
Go, we need ai.
00:05:05.254 --> 00:05:05.819
And you're like, oh, what for?
00:05:05.819 --> 00:05:09.512
I don't know, we just need ai, you know, and it's kind of like.
00:05:09.512 --> 00:05:14.286
I think it's a massive opportunity for consultants, by the way, unbelievable.
00:05:14.286 --> 00:05:17.252
But a consultant has to get really crisp on one.
00:05:17.252 --> 00:05:23.750
How do they uncover ai problems in an organization?
00:05:23.750 --> 00:05:27.860
And what I mean by that is is areas where ai can solve something.
00:05:28.583 --> 00:05:38.190
I heard this case the other day, which I and I love kind of hearing these ones that are unique from the, the field of sainsbury's right, which is a.
00:05:38.752 --> 00:05:55.091
For those of you not in the uk, sainsbury's is like a, a supermarket where you can go get food and stuff, and they are using ai to decide what meat they buy based on the weather conditions.
00:05:55.091 --> 00:06:23.922
For a sainsbury's store, because you think about it, if it's a heat, if it's hot and you know the uk's had a bit of a heat wave at the moment they know that sausages, burger patties and things like that are going to be in higher demand and and so when you think about this and stocking of a shelf in a supermarket, you can see that ai could be absolutely helpful in that scenario by analyzing where the patterns people are going to want to barbecue people are going to.
00:06:23.922 --> 00:06:26.471
You know they're not going to want, let's say, a stew.
00:06:26.471 --> 00:06:30.468
You know meat for a stew or mince or something like that, because they're going to be out and about.
00:06:30.468 --> 00:06:34.528
So let's stock the shelves with that because that product's going to be moved over.
00:06:34.528 --> 00:06:35.533
What might be.
00:06:35.533 --> 00:06:36.415
This is a time of year.
00:06:36.415 --> 00:06:39.480
We would traditionally put that on a full raw chicken.
00:06:39.680 --> 00:06:41.242
Yeah, right, yeah, exactly.
00:06:41.242 --> 00:06:49.507
That's a really good example and I find it really interesting because we didn't even talk about this before.
00:06:49.507 --> 00:06:55.911
But actually, sainsbury's is a model of data management.
00:06:55.911 --> 00:07:40.420
In 2016, 2017, if I remember correctly they were starting to have big problems with matching prices and, mainly, stocking up the shelves and making sure that their shelves have the best, the absolute best targeted audience, but at the same time, trying to be sensible, so you don't understanding the fact that you can't just put named brands, because that's where the profit is, and the fact that you need to, like, cater for everybody.
00:07:41.180 --> 00:07:54.271
So they did a huge machine learning exercise to understand how to stock their shelves in a way that makes sense, in a way that optimizes their storage units.
00:07:54.271 --> 00:08:07.896
They did a huge supply chain exercise where they would be able to offer prognosis over what they will have to buy, even in their storage areas.
00:08:07.896 --> 00:08:22.872
So what this tells us is that it still goes back down to data, because Sainsbury's is able to stock up their stores based on the weather projections in the UK.
00:08:22.872 --> 00:08:39.951
It does not mean that other supermarkets may be able to do it, because Sainsbury's already has the infrastructure and the platform to actually implementing it, and I think this is what potentially customers do not understand.
00:08:39.951 --> 00:08:47.134
Without that baseline, we can build a POC, that's not a problem.
00:08:49.485 --> 00:08:50.168
And then what happens?
00:08:52.102 --> 00:09:16.479
And with regards to how many people use AI tools, I remember at the beginning of last year where my best friend was very worried about artificial intelligence and how that's going to steal, you know, our jobs, and I was like, well, I'm not sure it's quite that way.
00:09:16.479 --> 00:09:28.347
And, yes, maybe some professions will be changing or disappearing completely, but others, you know, potentially will be enriched, whatever.
00:09:28.347 --> 00:09:37.067
I didn't have enough information to make a very good point either, especially because fear is a very, very powerful thing, right.
00:09:37.208 --> 00:09:37.811
Yeah, yeah.
00:09:38.659 --> 00:09:42.527
Well, fast forward to this year she has been.
00:09:42.527 --> 00:09:42.606
She.
00:09:42.606 --> 00:09:56.934
She's very smart individual, so she has been promoted twice and now she's in a management position and her number one job is to liaise with people.
00:09:56.934 --> 00:09:58.280
And what does that mean?
00:09:58.280 --> 00:09:59.780
Writing emails.
00:09:59.780 --> 00:10:04.302
And I'm like why don't you use AI to draft those damn emails?
00:10:04.302 --> 00:10:09.865
Because otherwise you're going to be working 14 hour days.
00:10:09.865 --> 00:10:12.527
You're never going to finish if you don't get some help.
00:10:12.527 --> 00:10:29.216
This is where I feel people in organizations can use AI, if they realize what's the real use case, not what they want to play with what is the real use case in their day-to-day lives.
00:10:29.775 --> 00:10:43.530
It's an interesting one, because AI you know a lot of what has been talked about is not a single thing, and one of the things that Benedict says in his article is it a product?
00:10:43.530 --> 00:10:45.485
What is it actually?
00:10:45.485 --> 00:10:53.966
And and because I've been using this metaphor that I feel like it is like electricity, and so therefore, it's an enabler of everything.
00:10:53.966 --> 00:10:56.561
It's an enabler of my toast that I eat for breakfast.
00:10:56.561 --> 00:11:00.490
It's an enabler of the you know, the fried bacon that I have.
00:11:00.490 --> 00:11:02.822
It's an enabler of the heat that I have in my house.
00:11:02.822 --> 00:11:04.283
It's it's an enabler of the heat that I have in my house.
00:11:04.283 --> 00:11:09.812
It's an enabler of so many things lighting, et cetera and so one, how do you ROI that?
00:11:09.912 --> 00:11:55.248
And I know Chris Huntingford, if he was with us, would have a lot to say on ROI and AI, and so I feel it is really a different paradigm, it's a different mindset and, like we've talked about the power platform, one of the things that have come up recently again for me is the, the difference between basic licensing and premium licensing, and this concept of of uh software economics, which is when you build a solution, and I was just yesterday talking to a company that has, I think it was a either 11 or 12,000 employees, 6,000 offices around the world and they are using the Power Platform on premium.
00:11:55.828 --> 00:12:07.634
You know for that organization and you know this is a corporate end customer using this technology, um, uh, out of Portugal.
00:12:07.634 --> 00:12:18.302
And what was incredible is that they said that the thing is, is that when we we did the ROI on the first project, right, it was a no brainer.
00:12:18.302 --> 00:12:23.347
But they said now every app we build from this point on, we're not having to think of who do we need to license?
00:12:23.347 --> 00:12:24.152
They're licensed already, right, the.
00:12:24.152 --> 00:12:24.895
Do we need to license?
00:12:24.895 --> 00:12:26.100
They're licensed already.
00:12:26.100 --> 00:12:27.563
Right, the right, the.
00:12:27.563 --> 00:12:28.525
Everybody's licensed.
00:12:28.546 --> 00:12:59.471
So now they're like well, the more solutions that we build that we architect now on this platform, we have diminishing costs as such, because it gets extrapolated right out across across that you build the economy at scale, and it's not just the amount of applications that you have to build, or flows or connectors or pages or whatever you're building, but it's also the supporting framework of it all.
00:12:59.471 --> 00:13:05.708
Right, you set up Android once you will always have to log into your applications.
00:13:05.708 --> 00:13:12.071
Right Done, you have put in that information.
00:13:12.071 --> 00:13:17.044
Once you will always be able to intelligently search for it.
00:13:17.044 --> 00:13:27.054
No matter what application that data sits in, no matter who touches it, you will be able to extract it back out.
00:13:27.054 --> 00:13:41.212
Our friend Chris said in his presentation in the latest presentation that I've seen of him, he was like you know how you like, type in data or import data into a system.
00:13:41.212 --> 00:13:45.331
The really cool thing would be to be able to find it again.
00:13:45.331 --> 00:13:47.413
How true is that?
00:13:47.774 --> 00:14:26.859
Yeah, yeah yeah, like that's so funny, but so true, right, and all of these things are things that you have probably built with your project managers and, I don't know, your BAs, your UX team, who know what to do always.
00:14:26.859 --> 00:14:27.666
It's.
00:14:27.666 --> 00:14:31.174
The scale is huge.
00:14:31.174 --> 00:14:32.157
It's humongous.
00:14:32.764 --> 00:14:59.379
What are you doing in the space of customers that have perhaps really leaned into the concept of personal productivity with a power platform and realizing that there's limits to how far you can go there and that a non-trained software developer builds software quite differently than a trained software developer?
00:14:59.379 --> 00:15:29.004
Right, and you know, standard rigor that we have around things like application lifecycle management, around testing in all its different formats, around documentation, supportability, all those type of robust things anything down to security-based testing of an application is what you have in a traditional software world.
00:15:29.004 --> 00:15:34.134
And then you've got the personal productivity and, dare I say it, citizen dev comes to mind.
00:15:34.134 --> 00:15:38.482
There's limits to how far one can go with that.
00:15:38.482 --> 00:15:40.067
What are you seeing around that?
00:15:40.128 --> 00:16:19.254
and moving into the area of more, um, you know, enterprise, like going up that ladder to into the enterprise more you know, in the last few months I've seen it a lot from from organizations who are starting, who have started rather straight on a mature platform, either on Dynamics or Power Platform, but with data first, to customers who have not, and they've leveraged SharePoint lists and Excel spreadsheets in order to create their productivity apps.
00:16:19.254 --> 00:16:25.352
They're all trying to navigate this world and, I'm not going to lie, it is not easy.
00:16:25.352 --> 00:16:27.493
It's like actually very complicated.
00:16:27.493 --> 00:16:40.078
From the organization who swallowed the citizen developer pill and they thought that actually we're just going to do all of this, all of this, ourselves.
00:16:40.078 --> 00:16:46.197
We only need a partner to support the stuff that we're doing.
00:16:46.304 --> 00:16:50.155
I know that you've said in previous iterations that the partner is dead.
00:16:50.785 --> 00:17:02.640
I would say that not yet, because, listen, these organizations they're not prepared for the level of assurance and diligence that the projects need.
00:17:03.384 --> 00:17:35.244
They're not ready to write, but they need to put in a lot of time to learn basic things like how to write a user story or how to ensure that it will travel correctly through all of your environments, how to make sure that you're not, you know, sending random emails to a production database.
00:17:35.244 --> 00:18:00.173
So that's like a huge, teething pain where I believe partners need to be consultants.
00:18:00.173 --> 00:18:18.012
You know Power Platform icon and started building applications and they're all hanging on SharePoint lists and Excel spreadsheets and some of them are like, oh, actually wait, we have a few licenses for you know premium Power Platform, but what do we do next?
00:18:18.012 --> 00:18:19.411
Who do we give them to?
00:18:19.411 --> 00:18:20.911
Do we buy more?
00:18:20.911 --> 00:18:35.845
They are truly recognizing the fact that they need the functionality, that they need the power, the discipline, the government, the governance, the security underneath it all, but it's very, very hard to get started.
00:18:35.845 --> 00:18:43.539
So that was a really, really, really long way of saying it's complicated to get started in these situations.
00:18:43.539 --> 00:18:45.645
Where do you think one should get started?
00:18:45.826 --> 00:19:03.119
actually, from your experience, yeah, well, when I ever think about where do you get started and it's not explicitly about what you're talking about I always think at the very beginning it's a very good place to start.
00:19:03.119 --> 00:19:08.888
When you sing, you begin with ABC, sorry, wrong show, this is a.
00:19:08.888 --> 00:19:10.294
It's not that story.
00:19:10.294 --> 00:19:12.482
No, I always think about data.
00:19:12.482 --> 00:19:13.246
Right, as in.
00:19:13.246 --> 00:19:18.487
I was brought up in that you model data up to the app layer, not the other way around.
00:19:18.487 --> 00:19:43.287
And I feel what my observation and I've worked with a few architects recently I am concerned about, about how even architects looking at building a solution from a data model perspective and I'm talking about going down to a field level, or is it a column level now, sorry, I still use the old nomenclature, um, because I was brought up with the old nomenclature.
00:19:43.287 --> 00:20:03.592
So the text field I hate it with a vengeance, right, yeah, and and the reason sorry, I don't hate it, like it's obviously has its place, but I hate it when it's used as a catch-all, right, and I see this, like you know, back back in in the early days of Dynamics CRM.
00:20:03.592 --> 00:20:18.232
There was a field in there, I just from memory and it said something like, let's say it was on a contact record and you said this contact was a manager and there was a field that said what's their executive assistant or no, it was their spouse, who was their spouse?
00:20:18.232 --> 00:20:19.615
And it was a text field.
00:20:19.615 --> 00:20:24.913
And I'm like, if it's another person, why isn't that a look up to a contact entity?
00:20:24.913 --> 00:20:28.123
Because now you leave intelligence in the system.
00:20:28.123 --> 00:20:32.596
Yeah, right, because you know a contact is more than a name.
00:20:32.596 --> 00:20:36.106
Right, it has other things, like an email address, a phone number.
00:20:36.106 --> 00:20:37.288
Why do I just want to know their name?
00:20:37.288 --> 00:20:39.733
Was it just so I drop it into a conversation?
00:20:39.733 --> 00:20:40.115
Anyhow?
00:20:40.115 --> 00:20:45.939
That aside, I see it on things like informed data collection online or anywhere where they go.
00:20:45.939 --> 00:21:13.098
The country field is a text field and you're like, imagine somebody reporting on that field now and all the spelling mistakes, um, that will come just with people typing the country name incorrectly, as opposed to that being you know a pre-populated list of what it could be, and then, of course, with longer lists, how do you make the interface so that you don't feel like you're scrolling through a massive drop-down list?
00:21:13.098 --> 00:21:20.191
But I'm like I feel that there needs to be more rigor around choosing the right type of field.
00:21:20.191 --> 00:21:21.955
Is it field or column?
00:21:21.955 --> 00:21:24.763
What am I meant to be saying I think it's column.
00:21:24.763 --> 00:21:26.028
Yeah, what type of column?
00:21:26.028 --> 00:21:26.189
What?
00:21:26.189 --> 00:21:34.837
What is the column type not name, but type that I should be selecting to make my system stay intelligent beyond just me building the application.
00:21:34.837 --> 00:21:36.221
Now, what is it going to mean from a reporting perspective?
00:21:36.221 --> 00:21:38.327
Because is it going to mean from a reporting perspective?
00:21:38.327 --> 00:21:45.288
Because it's going to wrap metadata around it based on, you know, the field, type, um, the, the column type you specify.
00:21:45.288 --> 00:21:47.894
Yeah, I just notice that there's.
00:21:49.516 --> 00:22:01.912
There's often not that level of rigor, yes to, and I always say the worst critic you can have is your customer, because your customer doesn't know what good looks like.
00:22:01.912 --> 00:22:05.826
Right, you can show them something and they might think, wow, that looks good.
00:22:05.826 --> 00:22:19.508
I always say the best critic is my competitor, yeah, somebody with the same skills as me in a different organization, because they're going to apply the rigor at the same knowledge level as I have to go.
00:22:19.508 --> 00:22:20.951
Is that best practice?
00:22:20.951 --> 00:22:23.019
Is that the best way of doing this?
00:22:23.019 --> 00:22:27.048
Is that going to make the system more intelligent or less intelligent, etc.
00:22:27.048 --> 00:22:27.450
Over time?
00:22:28.050 --> 00:22:32.666
And so I feel that there needs to be a robustness around that area.
00:22:32.666 --> 00:22:35.550
So when you say where do I start, I'm always thinking about you know.
00:22:35.550 --> 00:22:39.336
How's this data going to be used outside of just it's?
00:22:39.336 --> 00:22:42.299
Now, you know application, how is it?
00:22:42.299 --> 00:22:56.355
And that's why I love andrew's comment that the the power platform is not an app story, it's a data story, and I think, until people really start to realize that, and then and then my question is how do you learn to get good about data, how do you learn as an?
00:22:56.355 --> 00:23:07.417
What I mean is much more beyond just the power platform, but you know how to catalog it, reference it, how to make it available, how to sanitize it, how to, uh, archive it right.
00:23:07.417 --> 00:23:11.326
A lot of organizations have no archiving strategy, right?
00:23:12.048 --> 00:23:17.016
especially if you're young yeah, like if you started 15 years ago.
00:23:17.016 --> 00:23:22.933
You kind of know about data because you have to yes, there's no way around it.
00:23:22.933 --> 00:23:37.634
But if you didn't, if you're just trying to do something right now, it's really really hard to know about data and, even more tragic, it's really hard to show others the value of data.
00:23:37.634 --> 00:23:46.936
So I agree with you, we do start with data, we start with the data model, we start with you know, controls around it, guardrails.
00:23:46.936 --> 00:23:59.185
We start with the idea that we want to know more than that person's name, that we actually want to know her date of birth to send her a birthday card?
00:23:59.185 --> 00:23:59.946
I don't know.
00:24:00.528 --> 00:24:05.298
Her hair color, I don't care, but we need to know more about that.
00:24:05.298 --> 00:24:06.465
Yeah, so that's the idea.
00:24:06.465 --> 00:24:33.891
However, if I am the executive sponsoring this, in the age of AI, when I can see everything right now and instant gratification is even at higher levels, we need to come up with a way to make data sexy, to make it attractive, to make it the cherry on top, something that people want to see.
00:24:34.731 --> 00:24:43.682
Yeah, I think that's easier than what people realize to make it sexy.
00:24:43.682 --> 00:24:53.855
And the thing is that, you know, I remember on one project where the customer said for every item of data my salespeople put into the system, I want five benefits back.
00:24:53.855 --> 00:24:56.352
Right, and it really made you think, right.
00:24:56.705 --> 00:24:58.999
Yeah, right, benefits back, right so and it really made you think right, yeah, right.
00:24:58.999 --> 00:25:10.277
What you realize pretty quickly is that it's actually achievable, especially when you've got the ability to use a connector and bring in other data so you can enrich that data set by getting a small amount.
00:25:10.277 --> 00:25:21.133
So you know, for example, in the uk if you put somebody's postcode in, it narrows down to pretty much the house on the street even without knowing the street number.
00:25:21.133 --> 00:25:24.335
There's a high degree of accuracy which you can look up and validate.
00:25:24.335 --> 00:25:28.772
Things like phone number format Don't ask me to put in a certain format.
00:25:28.772 --> 00:25:33.997
I'll put it in whatever format that I think of, programmatically updated to the format that you want.
00:25:33.997 --> 00:25:43.758
Right, this stuff's all possible nowadays, but it's often that kind of polish and stuff that we leave off in the speed, I suppose, of getting things done.
00:25:43.758 --> 00:25:46.817
You know, don't make me enter the same piece of data twice.
00:25:46.817 --> 00:25:51.231
If it's this in the organization, bring it in from where it exists.
00:25:51.231 --> 00:25:58.000
Let's not create multiple copies um of or make people fill in data multiple times if it's available.
00:25:58.020 --> 00:26:02.276
Right, show it to me, it's one of my biggest pet peeves is that you know this information about me.
00:26:02.276 --> 00:26:04.231
You know whoever I'm buying from.
00:26:04.231 --> 00:26:06.069
Why are you asking me again?
00:26:06.069 --> 00:26:07.133
You know this?
00:26:07.133 --> 00:26:10.050
Yes, exactly, and that's where the rigor needs to come in.
00:26:10.644 --> 00:26:13.914
You know I just bought paint and that it was white.
00:26:13.914 --> 00:26:16.413
Stop trying to sell me white paint.
00:26:16.884 --> 00:26:18.932
Yeah, yeah, yeah, so true.
00:26:18.932 --> 00:26:35.576
Another thing I just want to share and I'll just bring up my screen here again is this comment, sorry, this slide that I saw this week from Satya, and I'm loving this right because back on our AI again.
00:26:35.576 --> 00:26:38.213
Ai is the greatest commercial opportunity in today's economy.
00:26:38.213 --> 00:26:44.892
This next generation of AI will reshape every software category in every business, including our own.
00:26:44.892 --> 00:26:56.096
From Satya, right, and what I'm seeing is that, and you can see here, bloomberg Intelligence 2023 is where the data on the right is referenced, and then the Microsoft Work Trend Index, 2024.
00:26:56.096 --> 00:27:02.955
But look at these numbers 280 billion an additional software revenue opportunity based on rising demand of ai products.
00:27:02.955 --> 00:27:06.671
71 say their companies use ai now.
00:27:06.671 --> 00:27:08.855
And I bring this up for a couple of reasons.
00:27:08.996 --> 00:27:16.949
One, the opportunity is massive right yeah you know, for every dollar invested, three to eight times the return and that that's really that 92.
00:27:16.949 --> 00:27:27.749
There there is the partner opportunity on this, and so there's a real need for partners to upskill their staff right, because you can't?
00:27:27.749 --> 00:27:31.837
And I feel there's two kind of skilling types that are needed.
00:27:31.837 --> 00:27:55.648
What you see on Microsoft Learn is how to use our technology skilling, and I see that the world and the community that we're in seems to have indexed heavily on that, but there's not a lot on like first principle, thinking around data outside of a piece of software as an example, right or or or any of these type of things.
00:27:55.648 --> 00:28:01.951
Like you know, you can technically get into what purview does, but what's the application?
00:28:01.951 --> 00:28:04.336
What's the broader implication of doing it?
00:28:04.336 --> 00:28:05.709
Why is it important?
00:28:05.709 --> 00:28:08.961
And it's answering all these business value questions.
00:28:08.961 --> 00:28:10.865
I think there's a lot more focus.
00:28:10.865 --> 00:28:16.653
The other thing is is that this number here of 71% say their companies use AI.
00:28:16.653 --> 00:28:33.936
I know, I feel I expect that's kind of bullshit, and why I say that is that the telemetry metrics that people use sometimes inflate these numbers or deceive these numbers, and it's happened for years.
00:28:33.936 --> 00:28:40.035
So, for example, if you took YouTube, tiktokiktok and facebook and you had a video on each.
00:28:40.035 --> 00:28:43.925
They will all count the number of views of that differently.
00:28:43.925 --> 00:28:48.496
So youtube, for example, say if it's watched for 30 seconds, that's a view.
00:28:48.496 --> 00:28:52.874
Tiktok will say well, if it loads on screen, that's a view, doesn't matter if it was played.
00:28:52.874 --> 00:28:54.467
If it loaded on screen, it's a view.
00:28:54.467 --> 00:28:56.410
Facebook might be five seconds.
00:28:56.410 --> 00:29:00.679
They all say this is my view, time right, and it's just sell a story.
00:29:00.798 --> 00:29:10.588
And and when we look at co-pilot, co-pilot, for example, you get the scenario that how do you tell whether you're getting adoption inside an organization?
00:29:10.588 --> 00:29:13.076
Is adoption measuring the right thing?
00:29:13.076 --> 00:29:15.644
So I I saw this example just yesterday.
00:29:15.644 --> 00:29:16.722
Where they're going, oh, look at this, we're measuring the word is the.
00:29:16.722 --> 00:29:16.972
So I saw this example just yesterday.
00:29:16.972 --> 00:29:28.954
Where they're going, oh, look at this, we're measuring Word is the most common copilot use case and I'm like hang on a second, how is it telling that somebody is using it in Word?
00:29:29.778 --> 00:29:32.125
Oh, a plugin fires when Word opens.
00:29:32.125 --> 00:29:33.326
It calls the copilot.
00:29:33.326 --> 00:29:40.412
So actually nobody used AI, just the software called and said you know, is the co-pilot connector alive?
00:29:40.412 --> 00:29:42.112
And we're counting that as use.
00:29:42.112 --> 00:29:50.405
Come on, nobody even used it, it just happened to be a trigger in the software and that's why I'm like these metrics sometimes, and it's.
00:29:50.405 --> 00:30:03.432
You know AI is a fuzzy term because AI has been around since the 60s, the 1960s, right, and it's like now, anything that's kind of an algorithm or a if, then statement, that's ai.
00:30:03.432 --> 00:30:13.767
You know that people are categorizing that as broadly ai as well, because the computer made a decision but it does make sense, like it does remember how we have.
00:30:14.426 --> 00:30:30.423
Um, what's the name of the app Office Vibe, I want to say, I don't know the one that tells you oh, you have collaborated with your peers for three hours a day, or you've had focused work for such and such.
00:30:30.423 --> 00:30:35.464
Well, that can definitely measure whether you're so.
00:30:35.464 --> 00:30:50.021
Are we measuring if someone has Teams Premium or Copilot for Microsoft 365, that they've used Copilot just because they've been in Teams or they did a Bing search or whatever they did?
00:30:50.021 --> 00:30:51.672
Because that wouldn't be accurate.
00:30:52.246 --> 00:30:52.707
Yeah accurate?
00:30:52.727 --> 00:30:54.791
yeah, if you look at it from one perspective.
00:30:54.791 --> 00:31:14.765
But if you look at it from your premises perspective, that says ai will be like electricity we will not feel it, we will not see it, it will just toast our toast brilliant then it's true yeah, valid, valid, you have a good point.
00:31:15.545 --> 00:31:18.747
and, and, and there is that whole thing around digital exhaust, right?
00:31:18.747 --> 00:31:38.096
All the stuff that goes on that you don't think about, which you know the graph is great at understanding and picking up and making sense of the information reported on might be more helpful in other contexts, even though that individual might not, you know, be doing it.
00:31:38.096 --> 00:31:43.900
But then, you know, will AI become so immersive that we're not on purpose using it?
00:31:43.900 --> 00:31:47.922
It's just like I don't think of, when I turn on a light switch that I am going to use electricity.
00:31:48.001 --> 00:31:48.883
No, I just lit the room.
00:31:50.724 --> 00:31:51.045
Yeah so I agree.
00:31:51.045 --> 00:31:56.548
And then shooting out a figure like 71% can make people believe.
00:31:56.548 --> 00:32:14.021
Wait a second, if this is 71% already, you know of organizations using it already why should I even bother to move my data into and model it into like a sensible program of work?
00:32:14.021 --> 00:32:14.982
Why should I do that?
00:32:14.982 --> 00:32:23.957
That looks like several millions of investment and it doesn't make sense because my stakeholders will not see it straight away.
00:32:23.957 --> 00:32:34.568
So maybe I should just ignore it and use AI the way I want to, because everybody is warning me that bad things are going to happen, but I've not seen them yet.
00:32:36.935 --> 00:32:37.136
Yeah.
00:32:38.386 --> 00:32:39.730
And I am worried about that.
00:32:39.730 --> 00:32:55.615
You know, in my day-to-day work I feel like instead of ramping up and investing and grown-up, problem solving organizations are taking corners already.
00:32:56.265 --> 00:33:08.230
Yeah, and I suppose in this probably we're almost, we're out of time, but yes, another thing that I find concerning is ai washing and yeah, what is?
00:33:08.269 --> 00:33:09.211
that tell me.
00:33:09.211 --> 00:33:10.755
So ai washing is this.
00:33:10.755 --> 00:33:24.134
You know, when I was in london, um, there was a, there was a pride event on and something that came out in the media was was, um, gay washing, uh, your business, right.
00:33:24.134 --> 00:33:32.715
And so businesses come out with their stance on, you know, inclusion and things like that and we're kind of riding the bandwagon.
00:33:32.715 --> 00:33:33.877
We've seen it in the.
00:33:33.877 --> 00:33:37.972
So it's kind of like we're seen to be doing the right thing, whether they're doing it or not.
00:33:37.972 --> 00:33:49.249
And we see the same thing in sustainability, right, in that whole scenario where we're ticking the boxes for the marketing, um, and then you see what's going on beside, behind the scenes.
00:33:49.249 --> 00:33:53.357
I've seen it in the whole area of women in the workplace.
00:33:53.357 --> 00:34:10.791
You know, oh, you know, we so appreciate equal opportunity and everything, and then women go on maternity leave and their job's gone when they come back, right, it's like the chairs got shifted and it's kind of like to the public, it's all, yes, we're this.
00:34:11.246 --> 00:34:14.293
Behind the scenes, it's like still cutthroat as ever.
00:34:14.293 --> 00:34:17.472
There's no adhesion to it.
00:34:17.472 --> 00:34:25.396
So when I look at it in the terms of AI, I see these companies now and all their marketing is AI, this, co-pilot, that AI, this.
00:34:25.396 --> 00:34:28.592
And I look at the company and go okay, how long have you been in business?
00:34:28.592 --> 00:34:37.717
15 years, oh so, and email's been one of the ones for me, right, like how do I get a real optimized email experience using AI?
00:34:37.717 --> 00:34:54.514
And I look at these companies that have got these AI tools and you're like no, you've been doing email classification for 15 years and all of a sudden, because AI has become the trendy word, you're now AI infused and it's kind of like it's like you know the whole.
00:34:54.514 --> 00:34:58.233
Responding to an RFX, it's like yes, we've got that feature.
00:34:58.233 --> 00:35:04.592
And it's like the customer's expecting a feature this big and you're meaning it's this big.
00:35:04.965 --> 00:35:26.500
But it ticks the box of saying, yes, it has AI in it, and that's why I think there's so much AI washing going on in the market, where people are calling their products, they're doing startups that are, they're riding the ai bandwagon, but there's no, you know, there's no substance to what's behind it.
00:35:26.500 --> 00:35:31.652
They're just using it as a marketing phrase rather than it's intelligent yeah, I see that.
00:35:31.771 --> 00:35:33.193
I see that a lot as well.
00:35:33.193 --> 00:35:39.280
I'm not sure I see it a lot in my customers, in our customers.
00:35:39.280 --> 00:36:01.222
I find my customers to be incredibly honest and very much set on good intentions Like I've never seen organizations really, really, really want to work for the greater good.
00:36:01.222 --> 00:36:08.338
However, I'm seeing a lot of partners go on and on and on with AI and what they do.
00:36:09.079 --> 00:36:09.260
Yeah.
00:36:09.605 --> 00:36:16.875
You know, with these tools for their customers, and I am wondering where are these customers who are doing extra?
00:36:16.875 --> 00:36:23.034
You know extra work in AI and it's also intelligent.
00:36:23.034 --> 00:36:25.478
Why aren't they in the news?
00:36:25.478 --> 00:36:27.931
And in fact, I did try to study this.
00:36:27.931 --> 00:36:46.739
I've listened to podcasts and I've read magazines and publications and some more obscure than others and the reality is nobody's offering any sort of consistent result that can hold water.
00:36:46.739 --> 00:37:01.376
You know that you can say yeah, yeah, okay, so I can see that organizations really are using AI, rather than 71% of organizations may have co-pilot in the corner of their being.
00:37:02.137 --> 00:37:03.681
Yeah, yeah, yeah.
00:37:03.681 --> 00:37:06.273
Well, that's all we've got time for today.
00:37:06.273 --> 00:37:11.056
Has anyone picked up the paid or the unpaid placement through the show?
00:37:11.056 --> 00:37:14.114
Can I have a look at your hands, anna?
00:37:14.114 --> 00:37:16.451
Are you also wearing a ring?
00:37:16.632 --> 00:37:17.996
that looks very similar to my ring.
00:37:18.786 --> 00:37:20.713
Oh my gosh, look at that.
00:37:22.085 --> 00:37:22.586
What's what?
00:37:22.648 --> 00:37:24.434
is what is this ring that we're wearing?
00:37:24.434 --> 00:37:26.541
Anna, it is, you sold me on it.
00:37:27.164 --> 00:37:30.030
Indeed, and we are like you know.
00:37:30.030 --> 00:37:38.371
My relationship with the aura ring um is a longer living than my relationship with my husband.
00:37:38.371 --> 00:37:47.159
Wow, I've had this ring since the very first generation of smart rings like that.
00:37:47.159 --> 00:37:52.074
It's got like little sensors that can tell me all sorts of things about myself.
00:37:52.074 --> 00:38:00.965
From how will I sleep about myself?
00:38:00.965 --> 00:38:03.672
From how will I sleep on whether I should do like some housework or go for a walk or do whatever, and it's like super inclusive.
00:38:03.672 --> 00:38:08.972
It doesn't want me to go on a spin class or something, but it does want me to move around.
00:38:08.972 --> 00:38:10.615
It tells me if I'm stressed.
00:38:10.615 --> 00:38:21.376
I've got like meditation techniques on my phone if I want to use them, but if I don't want to use them, it's smart enough that it doesn't push them on me.
00:38:21.376 --> 00:38:27.398
It's sort of like a smartwatch, but 10 times better, in my opinion.
00:38:27.398 --> 00:38:29.913
It's awesome, it's fabulous.
00:38:30.666 --> 00:38:46.590
I agree, I've had it now since Dynamics Mines or thereabouts, when Anna introduced it to me, and it's permanently on me and that's what I like about it over, you know, because I have an Apple Watch as well, but it needs to be charged so often and so it's not on me.
00:38:46.590 --> 00:38:55.032
But the beauty of the ring is that the only time it goes on the charger is when I'm in the shower, and that's enough, and it's like got about seven days charged.
00:38:55.032 --> 00:39:04.775
Now, this is not a paid placement, this is just us geeking out over our wearable technology.
00:39:04.775 --> 00:39:07.081
Um, and with that, anna, it's been a pleasure to just have us talking on the show.
00:39:07.101 --> 00:39:07.824
We don't need those other guys um.
00:39:07.844 --> 00:39:12.170
It was really lovely ciao, ciao, ciao, bye, bye.
00:39:12.170 --> 00:39:14.894
Thanks for tuningcosystem Show.
00:39:14.894 --> 00:39:20.931
We hope you found today's discussion insightful and thought-provoking, and maybe you had a laugh or two.
00:39:20.931 --> 00:39:26.952
Remember your feedback and challenges help us all grow, so don't hesitate to share your perspective.
00:39:26.952 --> 00:39:33.025
Stay connected with us for more innovative ideas and strategies to enhance your software estate.
00:39:33.025 --> 00:39:37.096
Until next time, keep pushing the boundaries and creating value.
00:39:37.096 --> 00:39:38.670
See you on the next episode.
SEARCH PAST EPISODES
Search past episodes of The Intelligence Age.
Disclaimer: The podcast and artwork embedded on this page are from Mark Smith [nz365guy], which is the property of its owner and not affiliated with or endorsed by Listen Notes, Inc.
EDIT
Thank you for helping to keep the podcast database up to date.
