July 30, 2024
Episode 331: In the Dynamics Corner Chair: AI is a Complex Concept and Confusion What It Can Do
This discussion focuses on AI and Copilot, aiming to address the confusion and misconceptions surrounding AI. Our guest, Kamil Karbowiak, emphasizes that AI is a complex concept and that more clarity is needed about its capabilities. The conversation also covers various copilots designed for specific tasks and domains. Kamil delves into the limitations and future potential of AI in Business Central.
Connect with Kamil on LinkedIn (https://www.linkedin.com/in/karbowiakkamil/)
#MSDyn365BC #BusinessCentral #BC #DynamicsCorner
Follow Kris and Brad for more content:
https://matalino.io/bio
https://bprendergast.bio.link/
00:01 - Understanding AI Co-Pilot in Dynamics
18:59 - Enhancing Conversational AI Understanding
22:39 - Navigating AI Permissions in Business
32:22 - Advancements and Ethics in AI
44:02 - Expanding Business Functionality With Copilot
59:49 - Innovation in Business Central With AI
WEBVTT
00:00:01.564 --> 00:00:07.650
Welcome everyone to another episode of Dynamics Corner, the podcast where we dive deep into all things Microsoft Dynamics.
00:00:07.650 --> 00:00:19.949
Whether you're a seasoned expert or just starting your journey into the world of Dynamics 365, this is your place to gain insights, learn new tricks and the possibilities of co-pilot AI in your life and business.
00:00:19.949 --> 00:00:21.666
I'm your co-host, chris.
00:00:22.560 --> 00:00:23.204
This is Brad.
00:00:23.204 --> 00:00:26.800
This episode was recorded on July 10th 2024.
00:00:26.800 --> 00:00:28.785
Chris, chris, chris.
00:00:28.785 --> 00:00:45.213
If I had a dime for every time I heard the word co-pilot, I would be retired yeah, right, we would just be doing this podcast for the rest of our lives my kids, kids, kids, which would be what great grandchildren podcast for the rest of our lives.
00:00:45.232 --> 00:00:49.289
My kids, kids, kids, which would be what Great grandchildren could probably live off of that.
00:00:49.289 --> 00:01:02.340
This world of AI is mystifying, and each time we talk to somebody about it, it becomes a little more demystified for me, so I have to keep having these conversations.
00:01:02.340 --> 00:01:07.352
Today, we had a great conversation with Camille Kawaiowski from Data Coverage.
00:01:16.781 --> 00:01:17.927
Good afternoon, Camille.
00:01:19.462 --> 00:01:21.549
Hi, good afternoon Chris, good afternoon Brad.
00:01:21.939 --> 00:01:22.641
Good afternoon, Camille.
00:01:22.641 --> 00:01:23.121
How are you doing?
00:01:23.121 --> 00:01:24.323
It's nice to talk with you again.
00:01:24.623 --> 00:01:26.686
Good afternoon Chris, good afternoon Brad, good afternoon Kamil.
00:01:26.686 --> 00:01:27.227
How are you doing?
00:01:27.227 --> 00:01:28.188
It's nice to talk with you again.
00:01:28.188 --> 00:01:29.271
It's very, very nice to talk to you guys.
00:01:29.271 --> 00:01:35.140
How's the weather and everything else in, wherever you are, it's hot, it's hot everywhere it's hot.
00:01:35.980 --> 00:01:36.924
It's hot everywhere.
00:01:36.924 --> 00:01:37.929
Over here we're baking.
00:01:37.929 --> 00:01:39.525
How is it over there for you?
00:01:40.021 --> 00:01:44.489
Same here, same here, well, in Celsius it's 35 degrees today.
00:01:44.489 --> 00:01:46.798
That Same here, same here, well, in Celsius it's 35 degrees today.
00:01:46.838 --> 00:01:47.200
Quite unusual.
00:01:47.200 --> 00:01:49.126
Yeah, it seems like it's hot everywhere.
00:01:49.126 --> 00:01:55.430
Everyone's cooking and everybody has a different interpretation of hot, but it seems like everybody's up in the heat.
00:01:55.430 --> 00:02:01.188
You've been running around, I see, so I appreciate you taking the time to talk with us.
00:02:01.188 --> 00:02:23.025
I've been wanting to catch up with you for a long time about a topic that we started to see a lot more traction on at directions north america back in april, and since then it seems like it's taking over the world and that is that is, ai, co-pilot and anything else related to that.
00:02:23.205 --> 00:02:39.191
It seems like you at directions or if we go back to then or even in between it's, it's almost difficult to go a day hour or maybe even a half hour without hearing some mention of the word co-pilot and AI.
00:02:40.501 --> 00:02:43.611
Everyone is getting good, getting a little bit tired.
00:02:47.360 --> 00:02:54.191
I'm getting a little tired of it too, and I'm getting tired of it for a number of reasons.
00:02:54.191 --> 00:03:24.592
But to try to get some clarification on this whole thing, I was wondering if you could talk with us a little bit about I know it's in your space and what you do within your realm of work and the services that you offer is hopefully shed some light and talk with us about some of the confusion around AI, co-pilot and what everybody thinks or what's considered over-promising for AI and what we can do.
00:03:24.592 --> 00:03:31.796
So before we get into that, can you tell everyone a little bit about yourself?
00:03:32.877 --> 00:04:21.512
Yeah, so Data Courage, camille Karbowiak, I'm the managing director of a company that is based in, headquartered in Poland, but we have offices and employees in different areas of the world, including US, and we focus on data analytics for the, you know, in very wide space, including fabric, including everything that is data warehousing and trying to get Power BI in the hands of people at various organizations, from the smaller to the ones that are in 17 or 20 countries, and trying to bring that up to speed for them.
00:04:22.420 --> 00:04:23.543
And AI has been with us.
00:04:23.543 --> 00:04:37.428
As we've met I think it was about five, seven, six years ago, something like that at the first conference, when Copilot was not existent yet and AI was taking shape.
00:04:37.428 --> 00:05:14.019
There was more machine learning at the time, and I think we've got a different, I would say, reiteration of that, with generative AI now adding another piece of complexity or actually something that was supposed to, in the natural language, answer any question that we might have around anything that happens around our data that we collect in our companies, and I think that is what we're also doing Now.
00:05:14.019 --> 00:05:19.451
We're helping also partners with arranging their story around AI.
00:05:19.451 --> 00:05:29.055
We're helping Business Central partners to deliver some things with things such as Copilot Studio, for example.
00:05:29.055 --> 00:05:40.654
So we do that as well, and we have our own products as well that we ship to the app source that are integrating directly with Business Central.
00:05:40.654 --> 00:05:45.449
So that is really what we do at Data Courage.
00:05:46.331 --> 00:05:47.733
Oh, great, Great.
00:05:47.733 --> 00:05:48.380
Thank you for that.
00:05:48.380 --> 00:05:50.326
You guys are doing a lot of great things.
00:05:50.326 --> 00:05:55.071
I see what you're doing and I also enjoyed the conversation we had and learning a little bit more about AI.
00:05:55.071 --> 00:06:03.230
Basically put in your terms, what is artificial intelligence and is it really intelligent?
00:06:08.180 --> 00:06:16.149
Complicated question it is it is, it is a very complicated question, and then you need to unravel those different.
00:06:16.149 --> 00:06:27.545
You know there's so much, as you mentioned, there's so much confusion around it and what is being put in front of us in a system or into an environment.
00:06:27.545 --> 00:06:52.370
It would be able to answer all these questions that we have around it.
00:06:52.370 --> 00:07:23.137
This is where I would like it to finally be, where you know, no matter what task you have for it, it will actually assist you, and I think we're not there yet and that's where the confusion came with a lot of the marketing that is coming from different channels, from different companies that it would be able to do that or it will be able to do that to an extent.
00:07:23.137 --> 00:07:46.339
You know, people think when they see certain videos or certain functionalities, they see that, oh, it can help me with so much more that is on that video and I wasn't able to do that before, and what we're seeing and experiencing is that it's not there yet in many of those spaces.
00:07:46.339 --> 00:07:59.627
So, if you think of the revelation of generative AI, there are four major aspects that generative AI is able to help us with.
00:07:59.627 --> 00:08:11.230
One of them is content generation, as we know that there is a content summarization as well, and there is also things like question and answer.
00:08:11.230 --> 00:08:16.011
An incarnation of that would be co-pilot chat, for example.
00:08:16.011 --> 00:08:30.894
And then there is the other thing that is also used and I think this is the broadest subject that has been picked up on by a lot of the partners who are kind of trying to get into this journey is the automation.
00:08:30.894 --> 00:08:39.389
So trying to automate certain tasks and these tasks that people want to accomplish.
00:08:39.389 --> 00:08:46.917
They can be assisted, ai assisted, but still it's a domain specific.
00:08:46.917 --> 00:09:18.259
It might be generic, such as things like bank reconciliation, but if you think about bank reconciliation aspect, it is a specific concept, it's around a specific functionality and it allows us generative AI is going to enable us, with specific prompting, to accomplish certain things and put some things into Business Central, into tables, and then have a human accept it or reject it or somehow correct it before posting.
00:09:18.259 --> 00:09:25.774
So it saves a lot of time and we cannot take it from the co-pilot functionality that it actually saves a lot of time.
00:09:25.774 --> 00:09:40.633
Even things like you know the aspects of Teams or you know trying to get notes from the meetings this really, you know Teams and Copilot for Teams, for example, is accomplishing really well.
00:09:40.633 --> 00:09:46.653
There are obviously other solutions to it, which has a different approach to that subject.
00:09:46.653 --> 00:10:06.349
But that's really the four domains where we think that the generative AI in our space, in our business central space, I would say, would help us and, of course, if you think about coding and things like that, that allows it to do that as well.
00:10:06.909 --> 00:10:21.311
Where the problem is is, you know and I've heard it many times over the conversations we had the directions with different people at different Microsoft sessions as well Is it like, is it one co-pilot or is it many co-pilots?
00:10:21.311 --> 00:10:53.253
And we know, right, we know that there are many of them, but this is just one of the aspects, and really the limitation of how this works now is that they do not communicate together yet, so these co-pilots one co-pilot cannot communicate with another one, pass some results to another one, and, of course, this is the journey that we're going through now.
00:10:53.253 --> 00:11:10.535
So you know, one of the things, one of the key takeaways, I guess, from this conversation, but also from the directions, is that there are many co-pilots and they're specific to different tasks rather than just one co-pilot for everything.
00:11:10.659 --> 00:11:12.287
That's what I wanted to ask.
00:11:12.287 --> 00:11:17.441
I have a lot of questions for you based on what you had just mentioned and hopefully I can.
00:11:17.441 --> 00:11:24.653
You know, I need co-pilot to help me, a co-pilot for my brain to help me go back and answer all these questions, or ask all these questions.
00:11:24.653 --> 00:11:33.481
You said there are many co-pilots.
00:11:33.481 --> 00:11:34.565
What is the difference between these co-pilots?
00:11:34.565 --> 00:11:37.575
Is it the data I guess, is that even the appropriate term the data that they work with?
00:11:37.575 --> 00:11:42.788
Is it the function that they perform or is it something else, like what is the?
00:11:42.788 --> 00:11:48.806
Again, it's like people saying I have a car or a vehicle, right, you have many different types of vehicles.
00:11:48.806 --> 00:11:55.133
You know I can differentiate between a sedan, truck, tractor, semi-tractor.
00:11:55.133 --> 00:11:56.767
You know there's a number of different things within there.
00:11:56.767 --> 00:12:01.392
When dealing with a co-pilot, what differentiates them from each other?
00:12:02.580 --> 00:12:03.746
So we need to go back answering this question.
00:12:03.746 --> 00:12:05.076
I think we need to go back answering this question.
00:12:05.076 --> 00:12:32.225
I think we need to go back to the, the whole large language model aspect and when you think about that, is that you have in front of you, uh, when you're using I don't know chat, gpt or whatever you're using the whole knowledge that it aggregated, and then you have to put it into a direction and give it more context of what you want to do and achieve with it.
00:12:32.225 --> 00:12:35.168
And the same goes with those different co-pilots.
00:12:35.168 --> 00:12:53.572
So if you have a co-pilot that helps you with giving text generation, marketing text generation, or a co-pilot that helps you with bank reconciliation, you're giving it a different context and giving it different prompts, right.
00:12:53.572 --> 00:12:57.630
So you need to tailor it to what you want it to do.
00:12:57.630 --> 00:13:05.390
It will answer questions in different ways when the domain changes.
00:13:05.390 --> 00:13:14.032
And that is really what kind of you know in the human nature we would say we would answer the same question differently to.
00:13:14.032 --> 00:13:21.542
You know, a five-year-old and someone who has a PhD, if we have that knowledge right to talk at that level.
00:13:21.542 --> 00:13:46.107
So that's why these co-pilots, or these agents or these assistants, they need to be taught, or I would say, their broad knowledge has to be kind of tailored to that specific task that we want it to accomplish, and that's why we had this.
00:13:46.980 --> 00:14:01.306
At the first we had this confusion, and I think that we're on the road of explaining it better, that we can actually plug it into different functionalities and start doing that.
00:14:01.419 --> 00:14:08.350
And then there's obviously the stuff that I've heard, also from Microsoft events directions as well.
00:14:08.350 --> 00:14:29.990
One of the sessions with Evgeny Korovin they were saying about they were putting things out there as in they will be introducing the connected co-pilots as well, so something that will perform a certain action and then that you know it's just like a function.
00:14:29.990 --> 00:14:39.407
If you think about it, it's a function that has certain tasks, it completes it and then it passes the parameters to a different one that is tailored to do something else.
00:14:39.407 --> 00:15:09.873
And we're doing this as well in one of our apps, for example, where we're passing certain things or certain things that one AI element is doing for us and then it passes it on to a different element and different model that has been pre-trained and fine-tuned to perform a different task and then, with the iterations of that, you have a specific functionality that you can accomplish and work with.
00:15:09.873 --> 00:15:12.096
You know and work with the results of it.
00:15:12.841 --> 00:15:26.863
I hear these terms and I do want to jump into more of some of the functionality that you can use with Business Central conversation, which I'll mention in a minute.
00:15:26.863 --> 00:15:31.581
But I do appreciate that AI could be used, or Copilot could be used, for, as you had mentioned, content generation or idea generation, which is helpful.
00:15:31.581 --> 00:15:41.462
I could do summarization of stuff right and we can talk about the stuff and then also you can ask questions, get answers.
00:15:41.462 --> 00:15:48.234
You had just also mentioned that you can train it right.
00:15:48.234 --> 00:15:55.337
I hear this word train almost like I'm teaching or training a person to do a job right.
00:15:55.337 --> 00:15:59.921
So again, now we go into some of the misconceptions and confusion and this might even take it a little bit deeper.
00:15:59.921 --> 00:16:05.755
You know, it might put us down another tangent what is it to train AI?
00:16:09.886 --> 00:16:16.484
I wouldn't say that you would train a new model for yourself, and I wouldn't want.
00:16:16.484 --> 00:16:42.605
I think we don't want to go that path At the level of and I'm limiting ourselves because there are some companies that have started doing and training their own models, but it takes a lot of cpu power, gpu power to do that.
00:16:42.605 --> 00:16:53.681
What we're doing is we are using a trained model and then we're using the pre-prompting techniques, so kind of something that also microsoft really puts on on their slides as well as grounding, yeah, so something that tells you how you should behave now.
00:16:53.681 --> 00:16:57.828
Well, how do you know in which domain we are now and what?
00:16:57.828 --> 00:17:09.807
What domain do we now uh want to focus and what we want your answers when they when, when it's specific to your, to that specific domain, yeah, so when we're in finance, we're going to be talking about finance.
00:17:09.827 --> 00:17:24.442
When it's finance for, I don't know, an NGO, or whether it's a finance for a company that sells electricals equipment, the answers should be tailored for that.
00:17:24.442 --> 00:17:26.723
So we're giving it more context.
00:17:26.723 --> 00:17:50.714
So we are not training the model, we are just adding, uh, the concept to, to train it based on, on data that uh, we, we, we would need to pretty much put a machinery and and start building it, start building the lm for a specific purpose, and I don't think that we should be uh, I don't know it's.
00:17:50.714 --> 00:17:55.288
It's a concept for these large companies that needs to solve specific.
00:17:57.884 --> 00:18:00.612
So is training in essence limiting?
00:18:00.612 --> 00:18:05.946
I'm trying to simplify this for my knowledge and my limited knowledge of AI.
00:18:05.946 --> 00:18:27.291
I know how to ask questions to co-pilot and get some ideas from it and create pictures and stuff, but is training in essence limiting the data that it reads, I guess you could say, or loads or associates or links or puts together to limit what it can do?
00:18:27.291 --> 00:18:42.030
Or is training it like teaching it a new set of functions, whereas if I have a picture, for example, and say, ai, this is a banana, now going forward, it can see that picture and it knows it's a banana.
00:18:42.030 --> 00:18:52.683
Or is it that it takes up all the data in the world, of all the fruits and vegetables, and it can see that someone already identified it as a banana and now puts it in there and I can say what is this?
00:18:52.683 --> 00:18:53.205
And it will already?
00:18:53.227 --> 00:18:55.423
know it's a banana, yeah.
00:18:55.423 --> 00:18:56.426
So training is that?
00:18:56.507 --> 00:18:58.528
yeah, Training is that machine learning?
00:18:58.528 --> 00:18:59.715
Yeah, it is.
00:18:59.715 --> 00:19:00.480
It's a different concept.
00:19:00.480 --> 00:19:02.287
Yeah, in that concept it's a different concept.
00:19:02.346 --> 00:19:02.989
Yeah, exactly.
00:19:03.300 --> 00:19:07.981
But you know that now LLMs are multimodal, so they also use the algorithms.
00:19:07.981 --> 00:19:10.846
That's even more complex because it's uh, you know it can.
00:19:10.846 --> 00:19:19.796
Now the multi-model that, yeah, the the term of multi-model is that you can show it, um, I don't know, some picture.
00:19:19.796 --> 00:19:20.982
It's going to generate a text.
00:19:20.982 --> 00:19:24.791
From the text, it's going to generate a picture or a video or something else.
00:19:24.791 --> 00:19:27.125
Yeah, from a picture it's going to generate a video.
00:19:27.125 --> 00:19:28.108
We've we've seen.
00:19:28.108 --> 00:19:29.721
I've seen that really great.
00:19:29.820 --> 00:19:39.471
Yes, you've seen sora, you've seen probably some other models that are out there and this is amazing how this is being done.
00:19:39.471 --> 00:19:44.310
And, of course, you know the problems with hands and so on and so forth.
00:19:44.310 --> 00:19:54.240
So, yes, we definitely have to probably go down a certain path in this conversation so that we are not getting lost in all of those concepts.
00:19:54.240 --> 00:20:19.127
So, yes, training with a specific data set, with training an LLM, is when you have a wide database of different concepts and then the LLM was trained with the vector database to identify which word is with proximity to another word.
00:20:19.127 --> 00:20:22.153
And I'm not saying it for no purpose.
00:20:22.153 --> 00:20:38.386
I'm saying it because I think that the next element that people have a misconception of and I think that you know we also discussed it with a lot of people why it doesn't handle data as it should right.
00:20:38.386 --> 00:20:42.122
So why it doesn't handle the numerical as it should right.
00:20:42.241 --> 00:20:43.987
Why is it hallucinating all the time?
00:20:43.987 --> 00:20:49.826
It can give you different answers to a very simple questions as multiplication.
00:20:49.826 --> 00:20:52.009
You know variances.
00:20:52.009 --> 00:20:55.914
It's not going to calculate it correctly, and why?
00:20:55.914 --> 00:21:02.231
And the reason is that it's how would you identify that one is greater than two?
00:21:02.231 --> 00:21:08.307
You need to somehow train a different type of model to be able to understand that.
00:21:08.307 --> 00:21:19.567
For the language model, it really is a very difficult concept to apprehend when you have a number somewhere, yeah, how do?
00:21:19.587 --> 00:21:20.368
you understand.
00:21:20.368 --> 00:21:22.073
Is it a small or is it a?
00:21:22.073 --> 00:21:25.946
We would need a math model, basically to explain how the numbers are related.
00:21:25.946 --> 00:21:32.006
I did learn about a vector and what it is and how it's there in a previous episode.
00:21:32.006 --> 00:21:35.553
It took me a long time to wrap my head around that whole process.
00:21:35.553 --> 00:21:39.550
By the way, so now I understand you use hallucination.
00:21:39.759 --> 00:21:50.647
I understand why it hallucinates now because a language model references information based on how it relates in space, not necessarily the function or use of it.
00:21:50.647 --> 00:21:52.986
See, I'm picking this up from what you're saying.
00:21:52.986 --> 00:21:54.001
Then tell me if I'm wrong.
00:21:54.001 --> 00:22:07.943
That's exactly right, and then if we wanted to use that information, such as math, we would need to teach it differently that they're not using these numbers as they relate.
00:22:07.943 --> 00:22:09.988
We're using them as a function.
00:22:09.988 --> 00:22:12.432
Man, man, I'm learning so much with these conversations.
00:22:12.834 --> 00:22:15.644
So yeah, it's conversational at this at the moment.
00:22:16.307 --> 00:22:20.642
Yes, yeah yes, exactly, that is a big thing with this and that's.
00:22:20.642 --> 00:22:30.393
I have conversations and it's almost like with anything new, and the negative bias that we have as humans is we come out and say it did this wrong, it did this wrong it.
00:22:30.393 --> 00:22:30.854
Look I can confuse it.
00:22:30.854 --> 00:22:31.695
Oh, look, I can have it go this wrong.
00:22:31.695 --> 00:22:32.057
It did this wrong.
00:22:32.057 --> 00:22:32.920
Look, I can confuse it.
00:22:32.920 --> 00:22:38.325
Look, I can have it go this way, when the reality is, as I say over and over again, it's just a tool that you use.
00:22:39.101 --> 00:22:50.170
So even if you're using it for content generation, as you mentioned, it may give you an idea and then you can use your content in your brain to expand on that idea, to create content.
00:22:50.170 --> 00:22:53.212
If you ask it questions, it could give you some ideas.
00:22:53.212 --> 00:22:57.345
You still should use it just as a tool, with the answers, and validate it.
00:22:57.345 --> 00:23:06.248
It will make you become, I say, more efficient, because it will give you ideas faster than if you had to do it on your own, just like a carpenter with a hammer.
00:23:06.248 --> 00:23:07.348
It's a tool.
00:23:07.348 --> 00:23:09.142
The hammer's not going to do anything on its own.
00:23:09.142 --> 00:23:10.945
It's all a matter of how you use it right.
00:23:10.945 --> 00:23:11.949
See, I'm getting there.
00:23:12.229 --> 00:23:13.913
Yeah, it's going to give you a baseline.
00:23:13.913 --> 00:23:34.912
It's basically that's where it's helpful, giving you a baseline, and then you take that information from a conversational standpoint and then the creativity still be coming from you, because you would still have to figure how would you use the result that AI gives you.
00:23:36.760 --> 00:24:13.893
So with that, let's take it back now within the space of Business Central and data, and how a business, a partner or someone can use AI within their application, or someone can use AI within their application, whether it be a partner creating an application or an extension for Business Central that utilizes AI, or whether it's the AI co-pilot stuff that Microsoft has added to Business Central and is planning to add to Business Central, and then this crazy thing, co-pilot Studio now, which again used to have another name.
00:24:13.980 --> 00:24:18.941
Everything seems to have a different name, and I'm just going to load this all on there because it's on my mind.
00:24:18.941 --> 00:24:43.575
Also, one of the concerns that I hear and I have as well, with ai in a business space not necessarily just business central, it's ai in a business space is the data, so one we have to train it to process our data or use our data, understand our data, which in an ERP system can be different because they have different modifications.
00:24:43.575 --> 00:24:58.324
But also I may have access to the entire system for data such as finance.
00:24:58.324 --> 00:25:00.829
Chris could work in another office in the company and he does not have access to finance data.
00:25:00.829 --> 00:25:01.813
He only has access to inventory.
00:25:01.813 --> 00:25:18.268
How can we ensure that if Chris uses the AI tool to ask questions, he doesn't get the finance data that I have access to, whereas if I ask a question, I get all encompassing item, finance and such.
00:25:20.372 --> 00:25:24.259
So we need to break it down a little bit, and I don't know.
00:25:24.259 --> 00:25:43.695
I'll start with the last one, because I think it's the concept that is also in the mind of, but the concept that is also in the mind of many and we already, I think you know, with one of the apps that we all of the apps actually, with Business Central, it's quite easy, right?
00:25:43.695 --> 00:25:57.825
So you are not giving the access, just like you don't give access to the chart of accounts, you are not giving the access to the functionality and the way that Copilot works is I think it's a I don't know how to say that, but I think this is a missing element.
00:25:57.825 --> 00:26:19.329
Yeah, with all due respect for Copilot and Microsoft, it's an element that is missing is that you should be able to pick and choose where you would want your Copilot to actually show up on your screen and have specific permissions for that person or attached to it.
00:26:19.329 --> 00:26:29.461
So the way we have approached it with the financial intelligence or any other app that we gave it, is based on the end user permission.
00:26:29.561 --> 00:26:47.611
So if the user has access to the chart of accounts, he's going to be able to ask questions and he's going to be able to use the AI financial intelligence, rather than having a copilot somewhere where it's a plugged in top of Business Central and you just start and chat about different things.
00:26:47.611 --> 00:27:13.922
And this is where I think that this has to change, so that people know that when they're enabling Copilot for finance or Copilot for bank reconciliation or any other thing that will be attached to, I know where this is coming from.
00:27:13.922 --> 00:27:30.930
You don't want a person from, for example, french area, for example, for French subsidiary, to have access to the German database or knowledge base.
00:27:30.930 --> 00:27:38.409
This has to be separated and if you ask a question, you need to have permissions to only ask questions about your P&L data.
00:27:38.409 --> 00:27:42.652
That is, in this company that you have access to.
00:27:44.288 --> 00:28:01.685
Yes, and that's where I see a big challenge because someone could have access to both, so they want to be able to analyze across both, and then, as you're building this vector of data, or we're teaching the system ai, whatever you'd like to call it how to analyze or process this data within its model for processing data.
00:28:01.826 --> 00:28:26.834
See, I'm trying to learn how does it know to put up the walls between people without changing the efficiency, because it's to me it would be a filtered data set for a person if they don't have permission at some sort of dimension, and then that dimension needs to be factored into the processing to eliminate a chunk of data.
00:28:26.834 --> 00:28:28.660
But it could be multi-dimensional.
00:28:28.660 --> 00:28:36.789
So it's, it's a big challenge and I don't, you know, I think somewhat, you just assume.
00:28:36.789 --> 00:28:39.347
Well, you just have to not try to understand it.
00:28:39.347 --> 00:28:42.282
But these are just some of the questions or concerns that I've heard.
00:28:42.282 --> 00:28:56.890
About Copilot, yes, it's great to create pictures, great to analyze information on the web, but now we want to transfer it to be a practical business solution not executive-wide, but business want to transfer it to be a practical business solution, not executive wide, but business wide, to gain the efficiencies.
00:28:58.381 --> 00:29:16.723
So, again, coming back to some of the customers that we have on financial intelligence and we were actually not that we were challenged by these questions, but people were just asking how it's going to behave and the way we have it.
00:29:16.723 --> 00:29:23.469
Now we don't have the, let's say, the consolidated version is ahead of us, so we're not yet doing the consolidation.
00:29:23.469 --> 00:29:45.968
But in terms of dimensions and whatever the user has access to, this is what we or a specific user can actually use within the analysis, right, so he's not going to be able to compare it with a different division or I don't know something else that is captured as dimension if he doesn't have access to it.
00:29:45.968 --> 00:29:48.728
Yes, there's a lot of work to do that.
00:29:48.728 --> 00:29:52.791
Yes, it will not give you those answers on that level because you don't have access and permissions to do that.
00:29:52.791 --> 00:29:57.214
Yes, it will not give you those answers on that level because you don't have access and permissions to do that.
00:29:57.214 --> 00:30:08.913
But on the level of the user that has access to all of these things, then it will pass on much more data.
00:30:08.913 --> 00:30:10.263
That is relevant.
00:30:10.304 --> 00:30:15.722
To answer a specific question about that, um, about that data set yeah, so it's all.
00:30:15.722 --> 00:30:17.205
Yeah, it starts with the data set.
00:30:17.205 --> 00:30:26.280
Then you, you give it some, uh, some information, some prompting, and then you're returning it based on the data set that you were given.
00:30:26.280 --> 00:30:33.294
Yeah, and how, how you're doing it in in the and how you prevent it from hallucinating is another aspect.
00:30:33.294 --> 00:30:47.932
But to have the permissions set and to control this, we were seeing that this is, you know, and Business Central actually helps with permissions to send the data.
00:30:48.900 --> 00:30:51.028
And then you know, and then there's another aspect of it.
00:30:51.028 --> 00:30:54.162
So people were asking oh, are you reading our prompts?
00:30:54.162 --> 00:30:56.688
Right, so is microsoft reading our prompts?
00:30:56.688 --> 00:31:07.942
Is microsoft knowing what's going out and in, given that that everything is in cloud, and why would you read the prompts and why would you make a deal out of it?
00:31:07.942 --> 00:31:11.828
And just the ai concept brings those questions.
00:31:11.868 --> 00:31:56.284
You know, there are people were very um, let's say um, they didn't want to go to cloud, for the same reason why they don't want to go to ai now, because it's the privacy, right and somebody reads what we're asking, uh, and, and what microsoft has, you know, announced several times, is that they, they do not see any of that because it passes through OpenAI and goes through that, and so that's the area of security, you know, besides the fact that you know there are terms and conditions that you accept with the co-pilot and whatever the.
00:31:56.284 --> 00:32:05.326
I don't know if there's any any way to check, uh, whether or not somebody is is is reading uh that.
00:32:05.326 --> 00:32:19.884
But we have already customers that are using our tools in production and, um, you know, they, they just want to have that ability to ask these questions to those numbers and I think that that's more valuable.
00:32:19.884 --> 00:32:30.344
Yes, it's a very difficult aspect to unravel without the people.
00:32:30.344 --> 00:32:34.708
That would say from the Microsoft side yes, we know, and we don't do that.
00:32:35.623 --> 00:32:37.308
We don't, because that's the thing.
00:32:37.821 --> 00:32:40.269
I think that they call it responsible AI.
00:32:40.269 --> 00:32:42.406
It's well written by them.
00:32:42.406 --> 00:32:44.287
I think it's been out for quite some time.
00:32:44.287 --> 00:32:48.443
I think it's what?
00:32:48.443 --> 00:33:00.346
Four rules or policies fairness, reliability and safety, privacy and security, which covers that, and then inclusiveness is their goal.
00:33:02.171 --> 00:33:02.592
Exactly.
00:33:02.592 --> 00:33:33.528
And then there's other aspects, because if you know I think it was today or the day before that the Copilot chat will not be available outside of US for until October, then if you read something like that, is it based on some regulations that were put in by, for example, I don't know European Union on those pieces, or whether it's the technology, what is behind it, that there's another, another thing.
00:33:33.528 --> 00:33:43.945
But if you cannot trust these kinds of you know big companies for for for being reliable in terms of your data, then we're not going to get anywhere.
00:33:43.945 --> 00:33:44.247
Yeah.
00:33:44.669 --> 00:33:45.210
And that's true.
00:33:46.461 --> 00:33:56.012
And then you know if, if you think of deploying your own model, you're going to be maybe I don't know two, three years time and you're going to be doing something for yourself.
00:33:56.012 --> 00:34:06.094
Yes, there are smaller LLMs that you can train and do it yourself, but then there's this whole cost of maintenance, trying to up-train it with new data and so on and so forth.
00:34:06.094 --> 00:34:10.231
It takes a lot of time and money to put it out there.
00:34:10.271 --> 00:34:10.672
That's true.
00:34:10.672 --> 00:34:12.085
I want to point out too.
00:00:01.564 --> 00:00:07.650
00:00:07.650 --> 00:00:19.949
00:00:19.949 --> 00:00:21.666
00:00:22.560 --> 00:00:23.204
00:00:23.204 --> 00:00:26.800
00:00:26.800 --> 00:00:28.785
00:00:28.785 --> 00:00:45.213
00:00:45.232 --> 00:00:49.289
00:00:49.289 --> 00:01:02.340
00:01:02.340 --> 00:01:07.352
00:01:16.781 --> 00:01:17.927
00:01:19.462 --> 00:01:21.549
00:01:21.939 --> 00:01:22.641
00:01:22.641 --> 00:01:23.121
00:01:23.121 --> 00:01:24.323
00:01:24.623 --> 00:01:26.686
00:01:26.686 --> 00:01:27.227
00:01:27.227 --> 00:01:28.188
00:01:28.188 --> 00:01:29.271
00:01:29.271 --> 00:01:35.140
00:01:35.980 --> 00:01:36.924
00:01:36.924 --> 00:01:37.929
00:01:37.929 --> 00:01:39.525
00:01:40.021 --> 00:01:44.489
00:01:44.489 --> 00:01:46.798
00:01:46.838 --> 00:01:47.200
00:01:47.200 --> 00:01:49.126
00:01:49.126 --> 00:01:55.430
00:01:55.430 --> 00:02:01.188
00:02:01.188 --> 00:02:23.025
00:02:23.205 --> 00:02:39.191
00:02:40.501 --> 00:02:43.611
00:02:47.360 --> 00:02:54.191
00:02:54.191 --> 00:03:24.592
00:03:24.592 --> 00:03:31.796
00:03:32.877 --> 00:04:21.512
00:04:22.420 --> 00:04:23.543
00:04:23.543 --> 00:04:37.428
00:04:37.428 --> 00:05:14.019
00:05:14.019 --> 00:05:19.451
00:05:19.451 --> 00:05:29.055
00:05:29.055 --> 00:05:40.654
00:05:40.654 --> 00:05:45.449
00:05:46.331 --> 00:05:47.733
00:05:47.733 --> 00:05:48.380
00:05:48.380 --> 00:05:50.326
00:05:50.326 --> 00:05:55.071
00:05:55.071 --> 00:06:03.230
00:06:08.180 --> 00:06:16.149
00:06:16.149 --> 00:06:27.545
00:06:27.545 --> 00:06:52.370
00:06:52.370 --> 00:07:23.137
00:07:23.137 --> 00:07:46.339
00:07:46.339 --> 00:07:59.627
00:07:59.627 --> 00:08:11.230
00:08:11.230 --> 00:08:16.011
00:08:16.011 --> 00:08:30.894
00:08:30.894 --> 00:08:39.389
00:08:39.389 --> 00:08:46.917
00:08:46.917 --> 00:09:18.259
00:09:18.259 --> 00:09:25.774
00:09:25.774 --> 00:09:40.633
00:09:40.633 --> 00:09:46.653
00:09:46.653 --> 00:10:06.349
00:10:06.909 --> 00:10:21.311
00:10:21.311 --> 00:10:53.253
00:10:53.253 --> 00:11:10.535
00:11:10.659 --> 00:11:12.287
00:11:12.287 --> 00:11:17.441
00:11:17.441 --> 00:11:24.653
00:11:24.653 --> 00:11:33.481
00:11:33.481 --> 00:11:34.565
00:11:34.565 --> 00:11:37.575
00:11:37.575 --> 00:11:42.788
00:11:42.788 --> 00:11:48.806
00:11:48.806 --> 00:11:55.133
00:11:55.133 --> 00:11:56.767
00:11:56.767 --> 00:12:01.392
00:12:02.580 --> 00:12:03.746
00:12:03.746 --> 00:12:05.076
00:12:05.076 --> 00:12:32.225
00:12:32.225 --> 00:12:35.168
00:12:35.168 --> 00:12:53.572
00:12:53.572 --> 00:12:57.630
00:12:57.630 --> 00:13:05.390
00:13:05.390 --> 00:13:14.032
00:13:14.032 --> 00:13:21.542
00:13:21.542 --> 00:13:46.107
00:13:46.980 --> 00:14:01.306
00:14:01.419 --> 00:14:08.350
00:14:08.350 --> 00:14:29.990
00:14:29.990 --> 00:14:39.407
00:14:39.407 --> 00:15:09.873
00:15:09.873 --> 00:15:12.096
00:15:12.841 --> 00:15:26.863
00:15:26.863 --> 00:15:31.581
00:15:31.581 --> 00:15:41.462
00:15:41.462 --> 00:15:48.234
00:15:48.234 --> 00:15:55.337
00:15:55.337 --> 00:15:59.921
00:15:59.921 --> 00:16:05.755
00:16:09.886 --> 00:16:16.484
00:16:16.484 --> 00:16:42.605
00:16:42.605 --> 00:16:53.681
00:16:53.681 --> 00:16:57.828
00:16:57.828 --> 00:17:09.807
00:17:09.827 --> 00:17:24.442
00:17:24.442 --> 00:17:26.723
00:17:26.723 --> 00:17:50.714
00:17:50.714 --> 00:17:55.288
00:17:57.884 --> 00:18:00.612
00:18:00.612 --> 00:18:05.946
00:18:05.946 --> 00:18:27.291
00:18:27.291 --> 00:18:42.030
00:18:42.030 --> 00:18:52.683
00:18:52.683 --> 00:18:53.205
00:18:53.227 --> 00:18:55.423
00:18:55.423 --> 00:18:56.426
00:18:56.507 --> 00:18:58.528
00:18:58.528 --> 00:18:59.715
00:18:59.715 --> 00:19:00.480
00:19:00.480 --> 00:19:02.287
00:19:02.346 --> 00:19:02.989
00:19:03.300 --> 00:19:07.981
00:19:07.981 --> 00:19:10.846
00:19:10.846 --> 00:19:19.796
00:19:19.796 --> 00:19:20.982
00:19:20.982 --> 00:19:24.791
00:19:24.791 --> 00:19:27.125
00:19:27.125 --> 00:19:28.108
00:19:28.108 --> 00:19:29.721
00:19:29.820 --> 00:19:39.471
00:19:39.471 --> 00:19:44.310
00:19:44.310 --> 00:19:54.240
00:19:54.240 --> 00:20:19.127
00:20:19.127 --> 00:20:22.153
00:20:22.153 --> 00:20:38.386
00:20:38.386 --> 00:20:42.122
00:20:42.241 --> 00:20:43.987
00:20:43.987 --> 00:20:49.826
00:20:49.826 --> 00:20:52.009
00:20:52.009 --> 00:20:55.914
00:20:55.914 --> 00:21:02.231
00:21:02.231 --> 00:21:08.307
00:21:08.307 --> 00:21:19.567
00:21:19.587 --> 00:21:20.368
00:21:20.368 --> 00:21:22.073
00:21:22.073 --> 00:21:25.946
00:21:25.946 --> 00:21:32.006
00:21:32.006 --> 00:21:35.553
00:21:35.553 --> 00:21:39.550
00:21:39.759 --> 00:21:50.647
00:21:50.647 --> 00:21:52.986
00:21:52.986 --> 00:21:54.001
00:21:54.001 --> 00:22:07.943
00:22:07.943 --> 00:22:09.988
00:22:09.988 --> 00:22:12.432
00:22:12.834 --> 00:22:15.644
00:22:16.307 --> 00:22:20.642
00:22:20.642 --> 00:22:30.393
00:22:30.393 --> 00:22:30.854
00:22:30.854 --> 00:22:31.695
00:22:31.695 --> 00:22:32.057
00:22:32.057 --> 00:22:32.920
00:22:32.920 --> 00:22:38.325
00:22:39.101 --> 00:22:50.170
00:22:50.170 --> 00:22:53.212
00:22:53.212 --> 00:22:57.345
00:22:57.345 --> 00:23:06.248
00:23:06.248 --> 00:23:07.348
00:23:07.348 --> 00:23:09.142
00:23:09.142 --> 00:23:10.945
00:23:10.945 --> 00:23:11.949
00:23:12.229 --> 00:23:13.913
00:23:13.913 --> 00:23:34.912
00:23:36.760 --> 00:24:13.893
00:24:13.980 --> 00:24:18.941
00:24:18.941 --> 00:24:43.575
00:24:43.575 --> 00:24:58.324
00:24:58.324 --> 00:25:00.829
00:25:00.829 --> 00:25:01.813
00:25:01.813 --> 00:25:18.268
00:25:20.372 --> 00:25:24.259
00:25:24.259 --> 00:25:43.695
00:25:43.695 --> 00:25:57.825
00:25:57.825 --> 00:26:19.329
00:26:19.329 --> 00:26:29.461
00:26:29.561 --> 00:26:47.611
00:26:47.611 --> 00:27:13.922
00:27:13.922 --> 00:27:30.930
00:27:30.930 --> 00:27:38.409
00:27:38.409 --> 00:27:42.652
00:27:44.288 --> 00:28:01.685
00:28:01.826 --> 00:28:26.834
00:28:26.834 --> 00:28:28.660
00:28:28.660 --> 00:28:36.789
00:28:36.789 --> 00:28:39.347
00:28:39.347 --> 00:28:42.282
00:28:42.282 --> 00:28:56.890
00:28:58.381 --> 00:29:16.723
00:29:16.723 --> 00:29:23.469
00:29:23.469 --> 00:29:45.968
00:29:45.968 --> 00:29:48.728
00:29:48.728 --> 00:29:52.791
00:29:52.791 --> 00:29:57.214
00:29:57.214 --> 00:30:08.913
00:30:08.913 --> 00:30:10.263
00:30:10.304 --> 00:30:15.722
00:30:15.722 --> 00:30:17.205
00:30:17.205 --> 00:30:26.280
00:30:26.280 --> 00:30:33.294
00:30:33.294 --> 00:30:47.932
00:30:48.900 --> 00:30:51.028
00:30:51.028 --> 00:30:54.162
00:30:54.162 --> 00:30:56.688
00:30:56.688 --> 00:31:07.942
00:31:07.942 --> 00:31:11.828
00:31:11.868 --> 00:31:56.284
00:31:56.284 --> 00:32:05.326
00:32:05.326 --> 00:32:19.884
00:32:19.884 --> 00:32:30.344
00:32:30.344 --> 00:32:34.708
00:32:35.623 --> 00:32:37.308
00:32:37.821 --> 00:32:40.269
00:32:40.269 --> 00:32:42.406
00:32:42.406 --> 00:32:44.287
00:32:44.287 --> 00:32:48.443
00:32:48.443 --> 00:33:00.346
00:33:02.171 --> 00:33:02.592
00:33:02.592 --> 00:33:33.528
00:33:33.528 --> 00:33:43.945
00:33:43.945 --> 00:33:44.247
00:33:44.669 --> 00:33:45.210
00:33:46.461 --> 00:33:56.012
00:33:56.012 --> 00:34:06.094
00:34:06.094 --> 00:34:10.231
00:34:10.271 --> 00:34:10.672
00:34:10.672 --> 00:34:12.085