Episode transcript The transcript is generated automatically by Podscribe, Sonix, Otter and other electronic transcription services.
Hello everyone. Welcome to the 5 Minutes Podcast. Today I like to discuss with you about metrics in AI projects and why I'm saying this. Today, every single company, every single project management office, and every single professional wants to implement AI projects. And this is absolutely fair. And it's most of the time the right thing to do because it's a super big trend, and it's something that I'm talking about absolutely all the time. However, FOMO, or fear of missing out, is not a beneficial criterion for implementing this type of project. Most of the companies and the people I see implementing this are not, I would say, putting down the numbers to make a rational decision on how they implement projects. Most of them just implement, thinking that all the costs are free and the benefits are unlimited. And this is not true. If we go to the cost side, we need to understand that artificial intelligence and, specifically, generative AI are not cheap resources. First, professionals working with this are professionals with competence. I'm not saying professionals that say that they are competent. I'm talking about people with real, actual experience in generative AI. They are not cheap. They are on the other side. They are probably one of the most expensive resources in technology and information technology right now.
The second is the cost of the processing machine. For example, let's suppose you use, um, Microsoft Azure, OpenAI service, or you are using OpenAI APIs. These resources, please. They are not free. They seem to be cheap because when you see, okay, 1000 tokens, it's, you know, it's less than a cent, and you think, oh, this is pretty much nothing. No, you are not right. Because when you use this on a scale, these costs become quite relevant. I know companies that are struggling because they are spending a lot of money on these initiatives. Most of the time, when you release them, you cannot remove them later because it will look very bad for you. So, most of the time, the expenditures for your project are very clear and not cheap. You need to work on the other side. It's what is the benefit? Where do I win with this initiative? And this is exactly what we need. For example, if you are just as an example, if you are a project management office and you are implementing tools, what you need to understand, okay, you need to do some simulations, and you need to see, okay, let's just use this as an example. How much time do I spend to generate, for example, a financial report of my project portfolio using the tools I normally have in my company?
How much time will I use if I use generative AI and this kind of tool or this kind of automation, or I would say a process on AI. Okay, it's half of the time. Okay. How much money does this mean? Is it worth paying for that? And I'm just challenging this because, look, I need to be honest. I'm working with this pretty much all the time today. And many times when I'm discussing with a client, I try to realize that and say, does this make sense to you? Sometimes, you can start small and evaluate the benefits. You can create a pilot and test it, and then you will see, okay, it's worth it. And then you scale up. I'm talking more on the enterprise level. When you are using AI on an enterprise level, you need to understand this because otherwise, you will spend a fortune just because you want to be together with everyone. And then you don't realize that, in mathematical terms, this is not the best thing to do. So always think about that. I hope you enjoy this podcast, and see you next week with another 5 Minutes Podcast.
