Day 0 Event #183 What Mature Organizations Do Differently for AI Success

Summary

This discussion focused on the adoption and maturation of artificial intelligence (AI) in organizations. Dr. Lamya Alomair began by providing a historical overview of AI development and its potential applications across various industries. She emphasized the importance of understanding AI’s core philosophy and starting with key projects to drive adoption.

Abdullah Alshamrani then presented findings from a survey of 650 organizations implementing AI. He highlighted that only 10% of surveyed organizations were considered AI-mature. The discussion revealed that generative AI has significantly impacted overall AI adoption, pushing organizations to upskill staff, increase AI implementation, and focus on AI governance.

Key challenges in AI adoption were discussed, including lack of trust, business alignment issues, and difficulties in demonstrating AI value. The speakers emphasized that 52% of AI projects never reach production, and piloting an AI project takes an average of 8 months.

Four main lessons for successful AI adoption were presented: implementing a scalable AI operating model, focusing on AI engineering practices, prioritizing upskilling and change management, and establishing robust AI governance. The speakers stressed the importance of building strong foundations rather than chasing the latest AI trends.

The discussion concluded with advice on improving self-confidence in AI development and the need for continuous upskilling to keep pace with rapid advancements in the field. Overall, the speakers emphasized the critical importance of a systematic, foundational approach to AI adoption for organizations aiming to become AI-enabled across multiple business units.

Keypoints

Major discussion points:

– Overview of AI history and development

– Impact of generative AI on overall AI adoption in organizations

– Challenges in implementing AI, including high failure rates and long pilot times

– Key practices of AI-mature organizations, including hybrid operating models, AI engineering, upskilling, and governance

– Importance of focusing on foundational elements rather than “shiny objects” when adopting AI

Overall purpose:

The goal of this discussion was to provide an overview of AI development and adoption in organizations, highlighting both the opportunities and challenges. The speakers aimed to share insights on how to successfully implement AI based on surveys of mature AI organizations.

Tone:

The overall tone was informative and instructional. The speakers maintained a professional, authoritative tone throughout as they presented research findings and recommendations. There was a slight shift to a more conversational tone during the Q&A at the end, but it remained largely educational in nature.

Speakers

– Abdullah Alshamrani

Role: Executive partner at Gartner

Expertise: AI adoption, organizational AI maturity

– Dr. Lamya Alomair

Role: Acting on the WT ship of technology foresight and digital economy at NCIT

Expertise: AI, technology foresight, digital economy

– Audience

Role: Attendees asking questions

Additional speakers:

– Dr. Laila Umayyad

Role: Co-host from NCIT

Expertise: Not specified

Expanded Summary of AI Adoption and Maturity Discussion

Introduction:

This discussion featured Abdullah Alshamrani, Executive Partner at Gartner, and Dr. Lamya Alomair, Acting on the leadership of technology foresight and digital economy at NCIT. The speakers provided insights into AI development, its impact on businesses, challenges in implementation, and best practices for successful adoption.

Historical Context and Foundations of AI:

Dr. Lamya Alomair presented a comprehensive historical overview of AI development from the 1950s to the present. She highlighted key milestones such as Alan Turing’s test, the first robot at General Motors, and the recent development of ChatGPT. Dr. Alomair emphasized the importance of understanding AI’s core philosophy, explaining that cognitive systems form the basis of AI, with machine learning and deep learning as subsets. She stressed the significance of AI in improving human cognitive abilities, framing it as a transformative technology that enhances our thinking capabilities.

Dr. Alomair also discussed the importance of use cases in AI adoption, presenting a formula for successful AI application: Use Case + Data + AI Algorithm = Value. She emphasized that organizations should focus on identifying valuable use cases before implementing AI solutions.

Impact of Generative AI:

Abdullah Alshamrani presented findings from a survey of 650 organizations implementing AI. He highlighted the significant impact of generative AI on overall AI adoption, noting that it has led to:

1. A rapid rise to prominence in organizations

2. Increased focus on AI upskilling and governance

3. A doubling of AI adoption in organizations over the past two years

Alshamrani provided specific statistics, including that 45% of organizations have adopted or plan to adopt generative AI within the next 12 months. This surge in interest has pushed organizations to upskill staff, increase AI implementation, and focus on AI governance.

Challenges in AI Adoption:

The speakers identified several key challenges in AI adoption:

1. High failure rate: 52% of AI projects never reach production

2. Long pilot times: An average of 8 months to pilot an AI project

3. Lack of trust in AI systems

4. Business alignment issues

5. Difficulties in demonstrating AI value

6. Data quality problems

7. Technical implementation challenges

8. Cost considerations, particularly for generative AI

Alshamrani emphasized the significance of the 52% failure rate, noting the substantial resources invested in AI projects that ultimately fail to deliver value. He also highlighted that only 54% of AI projects move from pilot to production, indicating a significant gap in successful implementation.

Best Practices for AI Maturity:

The discussion revealed that only 10% of surveyed organizations were considered AI-mature, applying AI across multiple business units and processes. Alshamrani presented four main lessons for successful AI adoption:

1. Implementing a scalable AI operating model:

– Adopt a hybrid model with centralized capabilities working alongside business units

– Develop reusable AI components and design patterns

2. Focusing on AI engineering practices:

– Establish robust pipelines for AI development and deployment

– Manage AI projects efficiently from conception to production

– Create reusable components to accelerate AI development and reduce costs

3. Prioritizing upskilling and change management:

– Implement AI literacy programs across all organizational levels

– Focus on both technical AI skills and broader AI understanding

4. Establishing robust AI governance:

– Address trust, risk, and security concerns

– Develop comprehensive frameworks for responsible AI use

– Include elements such as AI principles, risk assessment, and ongoing monitoring

Alshamrani stressed the importance of building strong foundations rather than chasing the latest AI trends or “shiny objects”. This approach was deemed crucial for organizations aiming to become AI-enabled across multiple business units.

Future of AI and Continuous Learning:

Dr. Alomair emphasized the importance of staying updated with AI developments and the need for continuous upskilling in AI. This sentiment was echoed in an audience question about improving self-confidence in AI development. Dr. Alomair advised starting with small projects, collaborating with others, and continuously learning to build confidence in AI development skills.

Conclusion:

The discussion provided a comprehensive view of AI’s transformative potential while highlighting the practical challenges of implementation. It encouraged a balanced and nuanced understanding of AI’s role in business and society, emphasizing the need for a structured approach focused on foundational elements and realistic expectations. Both speakers stressed the importance of continuous learning, strategic implementation, and robust governance in achieving AI maturity and success.

Key Takeaways:

1. AI adoption has doubled in organizations over the past two years, driven significantly by generative AI.

2. Only about 10% of surveyed organizations are considered AI mature.

3. Major challenges in AI adoption include high failure rates, lack of trust, business alignment, and data quality issues.

4. Key practices for AI maturity include implementing a hybrid operating model, focusing on AI engineering, prioritizing upskilling, and establishing robust governance.

5. Organizations should focus on building strong foundations for AI rather than chasing the latest trends.

6. Continuous learning and upskilling are crucial for individuals and organizations to keep pace with AI advancements.

Abdullah Alshamrani: Okay. Abu Khalid, you’re done? Okay. Bismillahir Rahmanir Rahim. Assalamu alaikum. Assalamu alaikum. Assalamu alaikum. Good morning, everyone. It is my pleasure to, to IEF, of course, and to this session, and myself, a co-host here, Dr. Laila Umayyad from NCIT, acting on the WT ship of technology foresight and digital economy. We’ll be delivering very interesting, has the foundation of AI, as well as some of the qualitative and quantitative data related to what mature organization AI, or what AI mature organizations are doing in that space. By way of introduction, my name is Abdushamani, an executive partner in Gartner, and without further ado, I will hand over to Dr. Laila Umayyad, who has an immense experience in AI domain, and we’ll be going through the first section of this presentation. Go ahead, doctor.

Dr. Lamya Alomair: Thank you, Amir Abdullah. Now, I’m going to talk about today. I will start with the question, which is, do you think technology can help? And do you think AI can help? When I start this, I will talk about the industrial revolution. When the first industrial revolution comes, it helps us with the steam engine, people to move around. And then when the second industrial revolution comes, with the electricity, it helps people to see even at night. And the third industrial revolution, with the communication and electronics, it also helps people to communicate all over the globe. So all the three industrial revolutions help our physical ability. But with the fourth industrial revolution, there is one more ability that comes with AI. It improves our cognitive thinking. So here, what I’m saying, that with AI, our cognitive ability is moving. So before, always we said that from the history, you can understand what will happen in the future. So, I will just talk about very brief history about AI. So everyone think AI started very soon? No, AI started in 1950, okay? When a mathematical professor, Alan Turner, was thinking, if the computer can think like human. So this is the starting point, 1950. So he trained the computer with information, and also he bring a human, and then he asked a couple of questions for this computer, and a couple of these questions for a human, and then he give a judge the answer of these two individual, computer and a human. And then, guess what? This judge could not distinguish between which one come from a computer, and which one come from human. In this moment, Alan Turner said, okay, that means that computer can think like a human. Then in 1955, John McKenzie in a conference, he launched the name of AI. After that, moving to the first robot, which was in General Motors. So, in General Motors factory, they don’t want to touch the chemicals, so they start with the first robot. So in 1961. After that, I just want to introduce you to the grandmother of Chad DBT, it’s Aliza. In MIT, in 1965, they have Aliza, which is the first chatbot. It was simple, but 25 years later, we got our Chad DBT. So Aliza is the grandmother, what I say, for our chatbot now. And then, you can see like around 25 years, there is nothing moving in AI. And then in 1997, IBM have this program called DEEP2, which is this program, beat the chess champion in the chessboard. So since then, IBM, they start doing these changes and they start to know that, okay, let’s do another champion. In 1999, in the lab of MIT also, one professor, she had the first promotional robot. So they merged the AI with the robot, so they got the first promotional robot. After that, Google started doing the self-driving car. And just to let you know, how important is the data for this and how to be on it also. Very mature when you use it. When they use it, they train it for only white people. And then when there’s a black guy who want to drive this car, it did not drive because there is missing data. This is where ethics coming and maturity of the data. And then, maybe some of you know the Jeopardy game. In 2011, IBM, they got the, what’s the word? The program that beat the Jeopardy champion. Then they continue doing, Google, they do also the DeepMind, another program. In 2017, there is a breakthrough in machine learning for skin cancer and also retinitis. So they start also the field of health here. And then in 2020, OpenAI, they launched the beginning of the Jeopardy game, which was an NLP, a natural app, which is going to the six and try to find the magic. And everybody remember 21 when Corona hit us and we need a vaccine, we need a good vaccine. So using AI, that’s why we can get the vaccine quickly. In DeepMind, AlphaFold, the protein structure can be predicted very quickly with the protein structure. It was taking like 10 years to have the sequence of the protein translated using the biophysics and then have the structure and then try to find drug discovery. But AlphaFold, it predicts the structure of protein very quickly and then the discovery of drug discovery. Also, we’re very proud to say that in 23, we had Alam, which is from Zadaiya, which is an authority and also they are improving now. Saying all of this huge history, we want to predict the future. We want to see what this will take. And at the same time, we try to learn from the journey before for having a new future. That’s why today me and my colleague, Engineer Abdullah, will try to give you a sample how to be a mature in applying this. So everyone is asking, what is the definition of AI? There is plenty of definition. But let’s say here, it’s a cognitive system, as I mentioned in the beginning. Cognitive system is a natural or artificial system that’s connected together, taking all this information and data and translate it to an output that mean something, understand, something we can get information from it. Something that decision maker can make the decision. Underneath this whole information, the artificial intelligence, again, it is a program algorithm that’s connected together, that’s getting all this information and have something analyzing, coding, and everything. This AI, after we say cognitive, and then the artificial intelligence, after that, there is a machine learning, which is a subdivision of AI. The machine learning is how to teach the machine with all this information. So we get all that data, and then we get this machine learning. Underneath this machine learning is N and V, which is analysis. So in the machine learning, there is different deep learning also. It’s the same as machine learning, but it has a different layer. All of these have a support application who creates optimizing system, N and V is the name for the tractor, but all of these is dependent on one view to understand the problem of the piece of AI, which is the cognitive system. So all of this can be created, and we can see any user group of this AI, either this is one of these machine learning, or having two or three, or having two or three of this machine learning. So next, we will have a look at a little bit more information than what we are here, and we are looking forward to… I forget about AI, about the elements of AI, about how to start it, about how to mature it. So, when we go and see, okay, what is the prioritization that can be in the business field of the government here? We just give you an example here. Government entity, or operational excellency or accomplishment of the government. When we look to AI, there’s customer experience, which is how we make the best of the government, how we put the customer in different segments, how we can plan our brand in different ways. When we go to operational, this is where we have how to manage, how to maintain, how to maximize the trust that we have. When we see business elements planning, programming, this can help us. And then, look at the function of how we can align with different companies, different organizations, different governments, how we can monitor it, how we can follow the signals that happen. All this, that can AI help us, is again the question. The new business here, I know it will be having some, creating some new rules, new rules, new rules, new rules, creating some new jobs, removing some jobs, but really, the new business will be, if we study it very well, it will be having a great impact. So, again, the AI can help us. I will just try to explain one example today. So, if you’re asking us how we can have the best and the most quality, maybe the machine learning. I told you about the machine learning. The machine learning is the computer algorithm that you give it some data and it will learn. So, if we give the data, the machine learning can give us any alternative that we want. And then if we apply the business model, how we find the business model, that’s what we can take, okay? Using the pool-based system, to make it according to divide and see who is the business, by understanding the rhythm that we are. And then, how we can optimize the inventory and making the, optimizing it, even if it means, decrease and lower the cost, make the benefit of all the information. So, all the AI can help us, okay? Now, let’s see how the campus can be optimized. So, the most important point here, when I talk about how to optimize the production quality, because we are caring about quality, we know that using AI can increase the quality. So, how does it mean we buy, embed it, analyze it, then try it, and then we teach them what is the pieces that maybe need failure. We can’t get it out. This is 60% of the failure. And then, I wanna use it to, so how we can analyze it. There are 20,000 things that can be analyzed. For example, if you take your cost for maintenance, okay, and then, for one piece, and then there is another one, you would analyze it using AI, and then you need another piece of maintenance. So, if we see the time, we have to be prepared. It is just going and coming. So, using the AI will reduce the time that we need to be prepared. Quality of time in the website, worldwide, with the 50,000 and 140,000, 1,400, 1,400, 1,400, this is very important, you know? And then, also, we are trying to make our life easier. We don’t want that any medicine, any services, all this will reduce time, and it will, the quality of our life, the quality of life will be reduced. And, I’ll tell you one thing. What I learned from education and biology is that we need the right formula for having the correct, applied framework. It is not easy to apply AI. It’s very difficult. That’s why we say it’s very helpful. I love it. It’s very helpful. That’s why we have it as well. So, the use, and referring to the use case, we have to study use cases very well. Not any use cases, I mean, the use cases that give us value. We understand it now. And, then, it’s very, very important. So, I usually say, with the booger, booger virus, they don’t know how to use a booger mask. It is not that, it’s not that easy for them. So, after care, it’s very, very important. So, nobody can do jobs if they are not dealing with them. Then, the disease, the disease. I usually say, there is a tsunami of disease. Okay, but how do you deal with it? This is very important. So, it’s very important to know about it, to have it, so we understand. So, these three, and number four, is very important, but I did not know that it’s gonna make the AI, make these things, the first one is the link, the second one is not. Having the graphical part, all of these have to make that, people can’t apply it very well. So, then, we understand that we need the right tools. And, then, we have to have the right government experience. All of these cases are government experience that we need to govern our disease. So, this is the formula that we have to use. And, then, we’ve got to have a body to do this. So, AI, you have to be having this body. This body, the upper body. So, we have to find a body. He is a little bit of a body, and he is a little bit of a body. This is the ability, you can see, how this community can choose from zero to one, and all of our problems that we have to solve. So, this is bad, you know. We have to do this, and we have to find a solution to this. Find my, you know, this is my operation, and how is this? So, we don’t know, if I, for example, tell me, tell me which one is not, and we can use this one. This is an example of the, we can do that way. And, it can be applied in any industry, okay? You have to, so, going down here, you can see that, four point, yeah, industry, personalization is very important. What they mean by personalization? Everyone has different choices, different needs, needs, so, this is very important to work on it, in AI. All of these, the prices, the importance, all of this, all of this, all of this, I have to work on it first. I cannot work on this with the people in AI. I cannot work with anyone else. I will keep on building this in my mind, but I will work on it. So, it’s very important to concentrate where to start. Where to start, first of all, it’s very important to understand this, to understand the core philosophy, and also to have a place where to start, to start in the five important projects and then, we’ll continue. And now, I leave you with my colleague, Dr. Ibn Abdullah, who will continue this presentation, and I’ll wait at the end for any questions.

Abdullah Alshamrani: Thank you, doctor. So, hopefully, you’ve learned, the foundational AI techniques, which are not really enough to apply AI within your organization. You need the five elements that Dr. Lamia talked about, and we’ve done some analysis in the market. We went out and surveyed 650 organizations around the world who are adopting AI in general, who excel in applying the AI techniques and apply the AI practices within their organizations. And we tried to even filter out these 650 organizations to understand what are the common practices really, really make organizations and AI mature organizations. So, out of the 650 that we surveyed, only 10% ended up to be AI mature organizations. And we’ll go through a lot of the learnings and a lot of the teachings that we’ve gotten out of this survey. So, basically, what we’ve done is a practical and qualitative approach to understand what the organizations are doing in the domain of adopting AI. So, we need to understand one thing. Then AI, generative AI, ChadDBT, in particular, came out end of 22, it really made a big splash. It really moved the needle forward when it comes to general AI adoptions. Why? Because a lot of expectations, a lot of hype, and a lot of value received out of applying general AI within the organization settings in general. So this splash had some sort of fripple effect on the whole AI adoption in the organization. So basically, it did not focus only on gen AI, but it also touched every other general AI technique as has been highlighted by Dr. Lamia in the previous section. And you can understand the numbers that are being displayed in this. So we’ve done this survey almost every year for the past eight years. And gen AI was not really a technical or an AI technique that we’ve highlighted as top ones. However, during last year, it came out of the splash, and one of the impact is to move from no applications within organizations to the highest used AI technique within the organization. As you can see, it actually compared to every prominent and famous AI technique, whether it is related to machine learning, NLP, optimization techniques, and others. So it came number one within the use of organizations, and it had a ripple effect on everything else as has been noted previously. And to understand this a bit more, I’ve sliced the remaining sections of the presentation to three parts. The part where gen AI, or the part that demonstrate where gen AI has really made an impact in the overall AI adoption, then the part that talks about the challenges related to overall AI adoption, particularly gen AI in specific, and the last one is to dissect the common practices around AI adoption within the organizations. So the top three impacts that were generated from gen AI in particular was focusing on upskilling, AI upskilling in the business itself, in IT or AI savvy sort of staff and associates within the organization. So this is number one. And the second thing is pushing the needle when it comes to AI adoption to next level, right? So gen AI had a splash around AI adoption in general. So because gen AI is focused and can provide value in specific use cases, but it does not provide value in every single other case. You need to use a plethora of AI techniques similar to the ones that were described by Dr. Lamia and the top AI within the previous slide. But most importantly, and it became non-negotiable, the AI governance itself, because it deals with sensitive AI or sensitive data itself, and it has become the center of the universe when it comes to AI policies around the world. So here in Saudi, for example, we have the personal data protection law that governs, that has a lot of governance mechanism when it comes to data related to AI itself. Aside from that, we have also the AI outline or the gen AI outline or guidance from SADAIA itself to give insights on how you should use gen AI and AI in general responsibly within your organizations. And as you can see from these three items, I mean, they are intertwined. Each one of them will impact the other. So the AI adoption entails that you upskill your team. And the more AI adoption that you have within your organization, the more care that you need to do when it comes to AI governance itself. So you cannot deal with one of them in isolation from the other, because they are really intertwined and impacting each other. Sorry about the delay, but the clicker here is not really the best AI technology in the world. But basically AI has a chain reaction when it comes, or gen AI in particular has a chain reaction when it comes to overall AI adoption within the organization. You know, the number here is very interesting. Since two years ago, AI adoption has doubled, has really doubled. We’ve received, we’ve understood that AI adoption gone from 1x to 2x within the organization, applied in multiple data or multiple business units as well as multiple business processes. This tells you that AI is becoming more and more and more stream and organizations are serious about it. Okay, in 22 or 23, they’ve been piloting, but in 24, they have been scaling. And that scalability is related to the 2x that I have been talking about. And one of the things that increased the adoption of gen AI in particular was embedding gen AI capability within the general application landscape. So if you have a CRM, for example, it is not uncommon to see a gen AI sort of capability within the CRM to help the normal user to try and digest many of the processes related to a very complex CRM, let alone the data that is hosted within the CRM itself. So basically, gen AI became a forcing sort of function when it comes to AI adoption. In fact, you see it more used within embedded applications comparing to isolated or standalone chat GPT or gen AI functions. And everything in between is outside really the embedded application. So basically, if you want to really, or the key lesson that I’m trying to say here, if you want to really push the needle when it comes to gen AI in general, let alone AI adoption in specific, you need to think about adopting AI techniques within the general business applications in your organizations. And this is basically what we’ve seen. If you work in software engineering, for example, or you work on orchestrating enterprise applications, the enterprise application strategy has moved away from only being composable and only being reusable to include the embedded intelligence. And embedded intelligence is nothing but including the functionality of AI models, whether they’re gen or normal, within the enterprise application landscape itself, just to make them more powerful, more impactful, and easier to use for normal users. As I said in the beginning, you know, gen AI has really made a big splash when it comes to AI adoption in general. However, it really, you know, verified that we need to stand humble when it comes to the general AI adoption, and it is forcing us to mature, you know, to higher stages with the plethora of challenges that we’re facing in the journey itself. And this is, and you know, some of the challenges I will go through in that section. And before I go through, you know, the details of these challenges, I wanna display two, you know, very interesting data points. So you need to understand that the journey toward AI maturity is not easy. You know, it is very challenging. It is very, it continues to be complex as we stand today, and it is very costly to organizations. And you know, one of the things that we noticed from our survey this year is that AI projects, on average, never go to production. They don’t see the light at all. 52% is not an easy number. 52% is a very large number. You know, you take an AI project, you pilot it for, you know, a number of months, as we will see in the next slide, then you ended up in throwing it, why? Because you could not really realize a break-even sort of point when it comes to the cost and value of these, of running these AI models or scaling them later on in production. So 52, this means, you know, it’s like tossing a coin, right? You don’t know whether it will succeed or it will fail. And the other data point that I wanted to talk about is the length of piloting, you know, a normal AI project itself. So in the previous, you know, survey, 23, it used to be seven months. It’s gotten worse this year with Gen AI, which makes, you know, things or organizations in general more frightened to take part of Gen AI and general AI adoptions within their organizations. They need to understand that in order to succeed, you know, in their pilot, and in order to succeed in the scalable aspects when it comes to AI adoption, they need to focus on different things as we will see in the next section. So, 52 is prone to failure and 8 months just to pilot an AI project, at the end you don’t know whether it will go to production or not. revolution that came before it. You need to understand the AI technique and club it with the AI value or the use case value for the organization. Without maximizing the end result of that sort of formula, you will not be able to actually say that we are AI enabled organizations at all. As you can see here, you know, the barriers are very, you know, lack of trust when it comes to AI, lack of business alignment, lack of data, lack of confidence in technological aspects of AI, especially for those who come maybe from certain countries where cloud is not really enabled there, provided AI models and AI capabilities are better in cloud, you know, environment comparing to local environment. And definitely talent is a big thing, but most importantly estimating and demonstrating AI value is the biggest challenge for organizations and it goes hand-in-hand with the previous two data points that we described, you know, 52% fail and it takes eight months to roll out a project from pilot to production. And if we focus on gen AI in particular, you know, you will see commonality between the gen AI and common AI adoption in general, but you know, the top ones related to technical implementations and cost of running gen AI and this year we’ve seen more reliance on thin ops, for example, techniques, especially for those who are acquainted about cloud computing. I mean they need, you need thin ops capability in place just to help you, you know, navigate the costs related to AI consumption and usage and without a doubt talent became also one of the top gen AI and this is where, you know, the gen AI literacy or AI literacy programs come to resolve that sort of challenge as we will see in the next section. So as I said in the beginning, you know, we’ve surveyed 650 organizations, right, and we’ve learned many data points related to their practices when it comes to AI adoption. And we filtered out the high maturity organizations which came out to be around 9-10 percent. So out of 650, you only have 65 organizations that are mature when it comes to AI adoption in the world. And we’ve learned, you know, four main lessons when it comes to AI adoption and scaling AI adoption within the organization. But before we go through these four main lessons, I want to highlight what maturity actually means when it comes to our analysis to the data points that we collected. So basically, to establish the maturity definition, we need to understand that the common or the AI mature organizations applied AI across several business units and processes. So it’s not enough to roll out a fraud management solution or model within production. You need to focus on customer intimacy, operation, you need to focus on predictive and preventive sort of scenarios, and this is related to different business units, let alone different business processes itself. And you need to understand that maturity as well means that they have deployed more than five AI use cases. Most organizations are still in the beginning of when it comes to AI journey. They are still piloting only one use case. The mature organizations, they’ve gone beyond that and deployed already five AI use cases in production. And they’re not in the early stages of that deployment. These AI adoption models and use cases have stayed in production for at least two years in average, which should have generated a lot of value for the organizations that use them. So we, based on the analysis that we did for data, we understood that AI mature organizations focus on the bottom, as you will see in the layer of that paragraph or that diagram. So basically the bottom forms the foundation of AI mature organizations relating to the operating model that is being used to scale AI within the organization, related to AI engineering practices and pipelines that are, that should be put in place, and definitely upskilling because we’ve learned in the previous surveys talent is a major issue. So upskilling is one way to resolve that gap in the organizations as well as change management, and definitely the governance aspect which focus on trust and risk and security management in general. So the ones on the top, whether it is related to the AI use cases, AI trends, the next big thing, and the AI models that are open in flux when it comes to announcements, you know, this is something you should not be focusing on if you want to really mature your journey when it comes to AI. You need to focus on the foundational elements as we will go through in a moment. So as I said, you know, AI, you know, in the previous diagram, the ones on the top are really shiny and you need to shy away from shiny objects when it comes to AI, and really be laser focused when it comes to the foundational and fundamental components of AI that will help you to really scale and reach the five use cases, different business units, and business processes, you know, hopefully reach two years production when it comes to AI adoption. But if we if we talk about the scalable AI operating model, and I am conscious of the time, so I’ll try to speed up, you know, when it comes to the operating model, you know, scaling AI requires different AI operating model. I mean, previously central teams may have succeeded to maybe pilot AI in the organizations, but in order to really scale it, you need to think about hybrid model, where very central AI capabilities need to take place, but they work in tandem and in collaboration with other, you know, business units within the organizations in a very specific and governed sort of manner, while the budget, when it comes to AI, you need to think about that mature AI organizations, you know, distribute the AI budget across different sort of business projects instead of being concentrated on one or two projects. And this is one example of, you know, hybrid operating model. As I said, there are central, you know, team that has the central capabilities related, for example, to AI strategy, AI architecture, and, you know, some of the subject matter experts when it comes to AI domain, but you also have the edges, the business units, the business processes where many of the innovations are being adopted, and in order to adopt that, you need to upskill the team, as we will see in a moment. But there is no one-size-fits-all, so every hybrid model will differ from one organization to other, but at the same time, you need to shy away from centralizing everything in one domain and think about, you know, what makes sense when it comes to hybrid in your organization. And AI engineering, if you come from software engineering background, for example, you need to, you will definitely relate to what I’m saying, what I’m going to say now, because you will, in software engineering, you have the pipelines that helps you to, you know, develop and design specific components and rolling out in scale to production, right? Same with AI. I mean, you need to have a mechanism to help you manage AI design and deployment end-to-end within you know, in a very automated fashion in your organization. Without building these sort of fundamental components in organizations, you will not be able to scale, you know, every AI model that you adopt within the organization. So the main focus for AI engineering in general, dedicated AI team, well-matured AI organizations should double down in AI engineering capabilities and practices within the organizations, right? And you need to understand that these AI engineering practices and capabilities will help you even, you know, ready your data when it comes to AI. adoptions. Gen-AI or normal AI? Traditional AI using the different AI techniques that we talked about earlier. And again, you know, mature organization doubled down in AI engineering and you can see here the top AI engineering practices relating to testing, developing AI solutions, and deploying AI solutions. If you focus on the developing AI solutions you will relate to the composable and reusable sort of components that you need to lay down if you want to really become serious when it comes to AI adoption in your organizations. Now it is, it is even more difficult to get it done. Okay, so basically AI design patterns are very similar to software design, software engineering design patterns. You, if you don’t come from that field, they’re nothing but Lego blocks, right? Lego blocks you could connect them like this or like that to actually come up with a specific shape that you have in mind. But basically AI design patterns help you to bridge the different use cases with the right AI solution architecture that you have in, that you want to build for your organization. And I’ll have one example in the next slide. So this example relates to clopping, you know, what we call retriever research model or RAG, retrieval augmented generation, with the general, you know, gen AI large language model. So basically, retrieval augmented generation happens to serve multiple use cases within the organization, whether related to customer intimacy or operational excellence or other sort of scenario like employee productivity or others. But basically this, you know, component here could be reused across many other AI adoption techniques and scenarios within the organization and you could clop it with the right LLM model that you could employ for the specific use cases that we, you want to really scale in your organization. But basically, you know, build that as a Lego component or as a reusable sort of component will help you to adopt other use cases in your organizations. And the third sort of teaching from the survey that we’ve done is the focus on upscaling and change management. So upscaling alone, you know, focusing on the AI associated should not be your only concern. You need to think about how you could, you know, adopt AI literacy programs within the overall organizations that help, you know, every associate within the organization understand the capability of AI and how they could use AI in their context. And this is where gen AI literacy or AI literacy programs could help. But definitely change management and we will see in a moment, you know, how change management techniques and activities could help you maximize the value related to your organizations across different, you know, spectrums and domains. So as we see here from that case study, basically this, you know, case study or the lesson that I want to highlight in this case study is that you need to be systematic when it comes to upscaling your associates within the organization. I mean, you should not focus only on the pros who use and reuse and create AI models or become really, you know, strong prompt engineers, but also focus on the bigger sort of group that may not really need advanced capabilities when it comes to AI, but they need moderate sort of capability. But the general, you know, associate having online courses that could be distributed across will help you to really reach every single associate within the organization. And basically, you know, that layering sort of systematic approach when it comes to AI literacy or adopting AI in your organization is very important to reach out more people. And again, change management is very important and change management, you know, techniques will help you to maximize the business outcomes when it comes to cost-saving, risk management, customer experience, or even productivity for employees. Without the right change management applications, you will not be able to reach very high, you know, sort of impact when it comes to the different, you know, values when it comes to business outcomes. And the last thing that I want to focus on is AI governance in general. And basically, AI trust, risk, and security management is one of the frameworks that we often highlight in Gartner, and it basically focuses on the fact that governance is being applied by diverse roles. AI associates or AI savvy people will not help you to reach high AI governance. You need to think about the different dimensions and different food or different perspective that needs to be put in place by these diverse diversity of roles. And the budget authority, especially when it comes to AI privacy and security, is very important. I mean, they need to be owned by a central unit that helps you to adopt the governance mechanism that is related. And business impact, when it comes to breaching or enabling AI privacy, is very important. So again, I mean, the AI tourism and or AI framework will help you to apply the governance mechanism, utilizing different components in the AI tourism technologies, and connecting with AI systems and the organizational governance practices that you have in your organization. And the last thing, you know, I just want to highlight is that AI adoption phases of maturity. And often, we focus on what’s, you know, above the surface rather than, you know, build strong roots that will help you really mature with time and reuse these components as you go. And this is just, you know, a rub to everything that I said and everything that Dr. Lamia said, you know, the opportunity is big when it comes to AI adoption. And you need to pick the right AI models using the right formula that was described earlier, you know, and the four lessons that I’ve gone through, whether it is related to picking the right operating model, a hybrid in that case, and utilizing AI engineering and upskilling literacy and the literacy program, as well as investing in AI trust and security management in general. This is very important to really, really push the needle forward and become an AI mature organization. Thank you very much. We’ll stay around if you have any questions. Any questions? Yeah, go ahead. No, it’s fine, because the people on the web, they need to hear you, but I can’t hear you now. Go ahead. Well, basically, they focus on the next big thing. They focus on the shiny object rather than on the fundamentals. I mean, you need to really focus on the fundamentals and, you know, instill and root out, you know, the different foundational components in your organization. Without this, right, you may be successful in one AI use case. You may be successful in one AI specific business process or business unit, but you will not be able to really scale different AI models across the whole organization. So, basically, AI maturity, it relates to AI maturity. If you want to really become, you know, AI-enabled organization in different, you know, regional and different business units, you need the foundation. So, they do not focus on the foundation. Yeah, it’s fine, it’s fine, we can hear you.

Audience: How can we improve our self-confidence? Like, even for AI engineers, I have some friends who are doing some projects for me, how to use metamorphosis. So, they are basically developing AI to develop AI. So, how can we improve our self-confidence in AI?

Dr. Lamya Alomair: Very great question. Upscaling ourselves is very important, as I mentioned before. Your question is very great, that how we can upscale faster, that doesn’t happen. I am really glad that this question has come in. Actually, usually, when you look at what is happening, usually, I’m looking at what is next. So, just keep continuing what you’re doing. Yeah, I know. There is a book I was reading that says the future is faster than what you do. Really, the future is faster. So, we have to upscale ourselves in the area that we have.

Agreement Points

Importance of AI foundations and fundamentals

speakers

Dr. Lamya Alomair

Abdullah Alshamrani

arguments

History and Foundations of AI

Focusing on AI engineering practices and pipelines

summary

Both speakers emphasized the importance of understanding AI foundations and implementing fundamental practices for successful AI adoption.

Need for continuous learning and upskilling in AI

speakers

Dr. Lamya Alomair

Abdullah Alshamrani

arguments

Need for continuous upskilling in AI

Prioritizing upskilling and change management

summary

The speakers agreed on the critical need for ongoing learning and skill development in AI for both individuals and organizations.

Impact of generative AI on overall AI adoption

speakers

Dr. Lamya Alomair

Abdullah Alshamrani

arguments

Generative AI’s rapid rise to prominence in organizations

Increased focus on AI upskilling and governance

summary

Both speakers highlighted the significant impact of generative AI on accelerating overall AI adoption and increasing focus on related skills and governance.

Similar Viewpoints

Both speakers emphasized the technical foundations of AI, with Dr. Alomair focusing on cognitive systems and Alshamrani on engineering practices, highlighting the importance of understanding and implementing core AI concepts.

speakers

Dr. Lamya Alomair

Abdullah Alshamrani

arguments

Cognitive systems as the basis of AI

Focusing on AI engineering practices and pipelines

The speakers shared the view that continuous learning and adaptation are crucial in the rapidly evolving field of AI, emphasizing the need for ongoing skill development and change management.

speakers

Dr. Lamya Alomair

Abdullah Alshamrani

arguments

Importance of staying updated with AI developments

Prioritizing upskilling and change management

Unexpected Consensus

Challenges in AI adoption

speakers

Abdullah Alshamrani

Audience

arguments

High failure rate and long pilot times for AI projects

Building self-confidence in AI capabilities

explanation

While Alshamrani focused on organizational challenges in AI adoption, the audience question about self-confidence unexpectedly highlighted a personal dimension to these challenges, suggesting a broader consensus on the difficulties faced in AI implementation at both organizational and individual levels.

Overall Assessment

Summary

The main areas of agreement centered around the importance of AI foundations, continuous learning, the impact of generative AI, and the challenges in AI adoption.

Consensus level

There was a high level of consensus among the speakers on fundamental aspects of AI adoption and development. This consensus implies a shared understanding of the critical factors for successful AI implementation, which could guide future strategies and policies in AI development and adoption.

Different Viewpoints

Unexpected Differences

Overall Assessment

summary

No significant areas of disagreement were identified among the speakers.

difference_level

The level of disagreement appears to be minimal or non-existent. The speakers presented complementary information on AI adoption, challenges, and best practices without contradicting each other. This alignment in perspectives suggests a cohesive understanding of the topic at hand, which may contribute to a more unified approach to AI implementation and maturity in organizations.

Partial Agreements

Similar Viewpoints

Both speakers emphasized the technical foundations of AI, with Dr. Alomair focusing on cognitive systems and Alshamrani on engineering practices, highlighting the importance of understanding and implementing core AI concepts.

speakers

Dr. Lamya Alomair

Abdullah Alshamrani

arguments

Cognitive systems as the basis of AI

Focusing on AI engineering practices and pipelines

The speakers shared the view that continuous learning and adaptation are crucial in the rapidly evolving field of AI, emphasizing the need for ongoing skill development and change management.

speakers

Dr. Lamya Alomair

Abdullah Alshamrani

arguments

Importance of staying updated with AI developments

Prioritizing upskilling and change management

Key Takeaways

AI adoption has doubled in organizations over the past two years, with generative AI driving increased interest and implementation

Only about 10% of surveyed organizations are considered AI mature, applying AI across multiple business units and processes

Major challenges in AI adoption include lack of trust, business alignment, data quality issues, and difficulty demonstrating value

Key practices for AI maturity include implementing a hybrid operating model, focusing on AI engineering, prioritizing upskilling, and establishing robust governance

Organizations should focus on building strong foundations for AI rather than chasing the latest trends or ‘shiny objects’

Resolutions and Action Items

Organizations should implement AI literacy programs to upskill employees at all levels

Develop reusable AI components and design patterns to scale AI adoption across the organization

Establish a hybrid AI operating model with centralized capabilities working alongside business units

Invest in AI engineering practices and pipelines to manage AI development and deployment

Implement comprehensive AI governance frameworks addressing trust, risk, and security

Unresolved Issues

Specific strategies to reduce the high failure rate (52%) of AI projects

Methods to shorten the long pilot times (average 8 months) for AI initiatives

Detailed approaches for estimating and demonstrating AI value to overcome adoption barriers

Concrete steps for organizations to transition from piloting to scaling AI use cases

Suggested Compromises

Balancing centralized AI capabilities with distributed implementation across business units through a hybrid model

Focusing on both technical AI skills and broader AI literacy across the organization

Combining off-the-shelf AI models with custom solutions tailored to specific organizational needs

With the fourth industrial revolution, there is one more ability that comes with AI. It improves our cognitive thinking. So here, what I’m saying, that with AI, our cognitive ability is moving.

speaker

Dr. Lamya Alomair

reason

This comment frames AI as a transformative technology that enhances human cognitive abilities, positioning it as the next step in human progress.

impact

It set the tone for the discussion by emphasizing the profound impact of AI on human capabilities, leading to a deeper exploration of AI’s potential and challenges.

Gen AI, or the part that demonstrate where gen AI has really made an impact in the overall AI adoption, then the part that talks about the challenges related to overall AI adoption, particularly gen AI in specific, and the last one is to dissect the common practices around AI adoption within the organizations.

speaker

Abdullah Alshamrani

reason

This comment structured the discussion into key areas, providing a framework for understanding the complex landscape of AI adoption.

impact

It guided the flow of the presentation, allowing for a systematic exploration of AI’s impact, challenges, and best practices in organizational settings.

52% is not an easy number. 52% is a very large number. You know, you take an AI project, you pilot it for, you know, a number of months, as we will see in the next slide, then you ended up in throwing it, why? Because you could not really realize a break-even sort of point when it comes to the cost and value of these, of running these AI models or scaling them later on in production.

speaker

Abdullah Alshamrani

reason

This insight highlights the significant challenges in implementing AI projects successfully, challenging the notion that AI adoption is straightforward.

impact

It shifted the discussion towards a more realistic view of AI implementation, emphasizing the need for careful planning and value assessment in AI projects.

You need to focus on the foundational elements as we will go through in a moment. So as I said, you know, AI, you know, in the previous diagram, the ones on the top are really shiny and you need to shy away from shiny objects when it comes to AI, and really be laser focused when it comes to the foundational and fundamental components of AI that will help you to really scale and reach the five use cases, different business units, and business processes

speaker

Abdullah Alshamrani

reason

This comment emphasizes the importance of focusing on foundational elements rather than being distracted by trendy AI applications.

impact

It redirected the discussion towards practical considerations for successful AI implementation, encouraging a more grounded approach to AI adoption.

Overall Assessment

These key comments shaped the discussion by providing a comprehensive view of AI’s transformative potential, while also highlighting the practical challenges of implementation. The speakers moved the conversation from theoretical possibilities to concrete strategies for successful AI adoption in organizations. They emphasized the need for a structured approach, focusing on foundational elements and realistic expectations, rather than being swayed by hype. This balanced perspective encouraged a more nuanced understanding of AI’s role in business and society.

How can we improve our self-confidence in AI?

speaker

Audience member

explanation

This question addresses the psychological aspect of working with AI, which is important for AI engineers and practitioners to effectively develop and implement AI solutions.

How to upskill faster in AI?

speaker

Dr. Lamya Alomair (in response to audience question)

explanation

This is crucial for keeping pace with the rapidly evolving field of AI and ensuring professionals can adapt to new developments.

How to effectively implement AI literacy programs across an organization?

speaker

Abdullah Alshamrani

explanation

This is important for ensuring widespread understanding and adoption of AI technologies throughout an organization.

What are the best practices for developing a hybrid AI operating model?

speaker

Abdullah Alshamrani

explanation

Understanding how to balance centralized and distributed AI capabilities is crucial for scaling AI adoption across an organization.

How can organizations improve their AI engineering practices and pipelines?

speaker

Abdullah Alshamrani

explanation

This is essential for efficiently managing AI design and deployment at scale within organizations.

What are effective strategies for AI change management?

speaker

Abdullah Alshamrani

explanation

This is important for maximizing business outcomes and value realization from AI implementations.

How can organizations effectively implement AI governance frameworks?

speaker

Abdullah Alshamrani

explanation

This is crucial for managing AI trust, risk, and security across diverse roles and perspectives within an organization.