AI Strategy and Investment Planning in Commercial Real Estate with CBRE | CXOTalk #822

AI Strategy and Investment Planning in Commercial Real Estate with CBRE | CXOTalk #822

Show Video

Welcome to see CXOTalk Episode 822. We're  exploring strategies around AI investment at CBRE,   the largest commercial real estate  services company in the world. Our   guest is Satnam Singh, the company's  chief digital and technology officer.  Tell us about CBRE and tell us about your role. We've got about 115,000 employees and about more   than 500 offices in more than 100 countries. We serve 95% of the Fortune 100. My role,   Michael, is twofold. I serve as the Chief Digital  and Technology Officer for the advisory services,  

part of the CBRE, where my team and I, we focus  on digital and data strategies, applications and   solutions across brokerage sales and financing,  property management and the appraisal businesses.   I also lead the digital and tech for our marketing  organization across all segments of CBRE.  Can you give us context about real estate  and real estate services so that we can   understand your your strategy within that? At CBRE, we really look at the entire real   estate lifecycle, right? So, we look  at all the way from asset ID, you know,   investment decisions, property management. We look at operational workflows, you know,   how we can introduce AI within those operational  workflows to serve our clients better. We look   at predictive analytics as I talked about, you  know, especially as you look at all the data that   we collect, we collect about 39 billion data  points across 300 different data sources. So,  

there is a lot of opportunity, if you will, across  the real estate lifecycle where the amount of data   that we have, we could really bring a lot of both  analytics as well as AI to bear, if you will.  So the data aspect of it is very foundational. Absolutely. We have for example, I'll give you   are smart facility management solutions, right?  We have about 1 billion square foot of across   20,000 global workspace solution clients  that we have. We have building operations   and utilization data and highly integrated data.  Right. That that we utilize through A.I. and using  

that you know we started looking at actions  such as automated maintenance, if you will.  Similarly, on the unstructured side, we have  data through a lot of documents, right? I mean,   there's a lot of documents in commercial  real estate, right. Whether that be a lease   document rental, pitch decks, if you will. So  we have, you know, a large foundation of both  

structured and unstructured data. How does that data then feed into   the underlying strategy that you have for AI? One is how do we really think about AI at CBRE?  So, if you think about the digital roadmap, how do  we align that with strategic business priorities?   How do we think about the client outcomes? What  are our focus areas? What are the market dynamics,   if you will, that are important for us? So that's  the strategic business aligned. The second is how   do we get faster time to value right? What data  do we have data that we can use or data that we   can source through our partnerships, if you will? In addition, what capabilities do we have? What   capability abilities can we develop as well as  acquire again through partnerships or otherwise?   And then how do we operationalize faster versus  the competition? So that's the second pillar of   faster time to value, but it's part of the  digital roadmap. The third thing is scale.   Michael the size of the problem and the universe  talent, uber salary, if you will, of the problem. 

Size means, you know, it's easier to solve  things of the scale of one property or one lease,   if you will, right? But when you really  start thinking about building platforms,   if you will, for multi segment solutions or  operating a portfolio scale of buildings,   you've got to think about this differently.  And again, at CBRE, we think about the data   platform, and this is why we take a very  balanced and pragmatic approach, right?  As I talked about the strategic alignment, we  really have practical air use cases that we work   on with the business teams. Our approach is an eye  for the sake of the AI or innovation for the sake   of innovation. It's about really utilizing  AI and technology in practical situations.  And that's where we look at. What are the  opportunities for automation or the operating,   you know, operational workflows, if you will?  And then as I said, right on the scale side,   again, going back to the data platform,  having 39 billion data points across   300 sources really is a huge benefit for us. We’re describing briefly the use cases, so your   use cases then are aligned to specific business  goals or specific outcomes that are important   strategically for CBRE. I'm assuming correct  me if I'm wrong, that that's what's going on? 

In real estate, you have three things, right?  You have transactions with a lot of transactions.   You have assets which require maintenance  and you have assets that are, in essence,   investments. Right? So when you think about these  transactions, right, they have documents, right?  As I was saying, there are leases. For example,  you want to extract data from these leases,   such as square footage. The price for square  for the rent and so on. Second is, you know, as   I talked about maintenance, right, things such as  cleaning of property management, if you will. And  

the third thing is investments where, you know,  you think about lots of data coming together, both   structured and unstructured, and you have both  data at a macro level as well as insights that   you gather on a more local level, if you will. How do you think about AI and in ways that are   different than other kinds of technologies? you know, if you step back and you say, look,   there are many things that you can do  with structured data that necessarily   you can build predictive insights and  you can get predictive intelligence,   if you will, but it really comes in handy. And we need to step back and say, look, if I   as a flavor has been around for a while, right?  I mean, there's two aspects to feeling like this   is my personal way of really thinking about it,  which is that when we talk today about AI, we   generally talk about generally the AI or we think  about generally the AI, which is about, you know,   how do you develop content, new content such as  text images or video by giving a variety of input.  Now, previously we've also had the AI,  which again, my personal term, if you will,   for it is analytical, right? Where you've had  for a long time, you've had the structured data   that financial companies or companies in retail or  transportation have used, right? So for example,   if you think in retail personalization and  commerce, Right. Is there right. And you know,   there are companies look at, hey, how do I think  about the placement on messaging generative the AI   is about really looking at creation of that  documents or analyzing those documents. 

And again, when we look at it from a perspective  of a commercial real estate firm, right,   we look at one of the opportunities, if  you will, to create value, to create that   productive enhancement by looking at things such  as lease documents, if you will. We in our own   internal analysis, if you will, our internal  work that we've done, for example, looking at   manual review of lease documents, we've been  able to use AI to cut that manual time by 25%,   if you will, using the AI to extract that data. And similarly, there are other documents such as   inspection reports, if you will. And then  there is the other opportunity around,  

you know, smart being smarter about facilities  management, if you will. Right? So again,   it's the sum of the foundational aspects will  remain the same. Michael It's again about   contextualize and get right. How do you bring  that context, if you will, to the real estate  

workflows and the real estate operations, which,  by the way, is no different than taking a AI and   contextualizing it for any other industry. At the end of the day, you have to take a   platform, if you will, a large language model  or a predictive capability and make sure you   effectively contextualize it in your specific  workflows, in your specific industry, if you will.  When you say contextualize that, can you  elaborate on what you mean by that and what   are the pieces? How do you go about that? A lease document has a different kind   of data points than, let's say,  some other document we'll have.  Right. And so how do you understand through an LEM  model that you want to be able to extract specific   items such as the square footage, if you will,  or you want to extract something like the price   per square footage or you want to extract the  address, if you will, of the property that this   lease that's about right in there, you have to  make sure quote unquote, call it the right prompt,   the right entity that you're trying to extract. How do you contextualize that for real estate?  

Right. And that's what I mean by is that there is  a certain type of call, a dictionary, there is a   certain kind of call it entity, if you will, that  you are trying to recognize from that document,   if you will. Right. And you have to be you  have to think about the right set of prompts.  You have to think about the right set of  query quoting that document, if you will,   to be able to extract information that is relevant  to your operations as a real estate company.  Is it mostly around efficiency, the  improvements that you're looking for,   or are there other use cases as well? it's not just purely about efficiency,   right? It's again, when we look at this, we  look at it from a perspective of productivity.   Enhancements are definitely one of. That's right. And productivity enhancements   means we have to look at the workflows  that we have. Right. So for example,  

Michael 822nd episode of this podcast, right.  I'm sure you have a certain workflow that   you and I were talking about when before we  went live, right? You have a clearly defined   workflow and as you look within that workflow, you know what opportunities exist, if you will,   to optimize that workflow for your benefit, right? So similarly, we look at that from a productivity   perspective. In previous lives, I've had the  fortune of being in other industries such as   tax and insurance and others. Right. And again,  you look at the workflows from that productivity  

and has now actually give you an interesting  story. When I was in the music industry a while   back at a company called Snow Camp, which was  next Napster, second innings, if you will, right.  I found out that, you know, music companies  would actually call the the different stores   like Tower Records, if not a lot of people  maybe aware that there was a company like   that. But the analysts sitting at the music  label would actually call the Tower Records   all across the US and call them and say, how  many how many volume of this did you, Bobby CDs,   did you sell of this album, if you will? Well, now you can get that data with digital   distribution. That data is much easier to get.  So, what do you know, now you can focus on more  

value generating, more higher level activities,  if you will, insights that they had that analysts   can do. So that's definitely the productivity  announcements. The second thing is you look at   the operating model, right within your existing  business, what operating models can you improve   or can you transform business models, if you will? Right. And that's where he's really coming at it.   A slightly different example, if you will. But  in China, I was actually just reading up a few   days ago, very interesting. And in China, you  have these e-commerce marketplaces that have   a lot of influencers, right, selling goods. And  what they're doing now is you have these virtual   avatars of these influencers that can get ready  for about, you know, $1,000 or 1100 dollars. 

And they need just one minute of video. And these  virtual avatars are up at 3 a.m. in the morning,   4 a.m. in the morning, selling the goods.  And if you pay a little bit more, they will   actually read the livestream of comments  and reply to them. So very interesting,   you know, in terms of how AI is introducing  that opportunity for new operating models.  And then the third piece is around how  do you use AI? And we look at it to say,   how do we use A.I.? For my marketing leadership  or bringing new thought leadership to the market,   you know, things like sustainability or climate  tech, if you will, which a lot of you know,   our clients are interested in and are  baking into their forward-looking plans. 

Please subscribe to the CXOTalk newsletter.  Subscribe to our YouTube channel, leave   comments and check out  We have incredible shows coming up.  You mentioned earlier the issue of culture and  we've been talking about data and the role that   data plays and being able to drive these kinds  of changes. Where does culture fit into this?  You need to think about what I call the  digital roadmap and then strategic career. 

You need to think about your culture, if  you will. Right. And I think within culture,   there's a couple of aspects that stand out.  First and foremost is definitely, you know,   what I was talking about earlier is how do  you really think about that? The strategic   business alignment you really bring this, isn't  it? You know, it isn't just a technology tool. 

You know, if we're thinking of anybody who's  thinking about AI purely as a technology   answer isn't doing justice to the isn't thinking  about that the right way and is definitely not   approaching it the right way, because then they  are probably doing tech for the sake of tech or,   you know, innovation for the sake of innovation.  If you have a if you think you have a hammer,   then everything looks like a nail, as I said. Right. So that I think is the first and foremost   thing is how do you really think about  AI as an opportunity to solve something   differently? Right? You previously could  not solve it because you didn't have the   right without it. You didn't have the right  infrastructure or the complexity was so large,   right, that you can solve. So that's one is how  do you think about contextualizing it again within   that business, within your strategic alignment? That's one, I think. The second is creating a  

culture around experimentation, like with any new  technology, right, if you will, And I mean this   in the in the context of generative AI, if you  will, right? Not everything will be a slam dunk,   right? So what you have to do is create a  culture of experimentation about learning   fast and most importantly, moving at pace and  urgency so you can operationalize even faster.  Right? So that's the second pillar, if you  will. The third pillar in my mind is what I   call responsible innovation. It would be the one  thing you have to be mindful about careful about  

is the potential for hallucinations. Right. If  you ask the wrong question, if you will, that   is not properly contextualized, then you're going  to potentially get some wrong answers in there.  Which means that this is not a silver bullet  to everything. Right. You have to think about   validation. You have to think about you have  to think about validation of process with some   human in the loop, if you will. Right. So that's  what I mean by culture is right. You have to step  

back and say, How do I really get started,  if you will? How do I really operationalize?  So, you're thinking very explicitly then correct  me if I'm wrong about your processes and changing   your process needs to reflect the realities of AI. Like you said, like with large language models,   the potential for hallucinations. it's not about doing it for the sake of it,   but we're actively saying, look,  as we look at these workflows,   if you will. Right. What are the possible areas  for those API interventions in these workflows?  Right. And we you know, if you look at workflow  all the way from your identifying an asset, if you  

already or some kind of brokerage where there's  some leasing discussion going on, you're trying   to figure out, Hey, I have certain parameters  for which, you know, that are related to this   asset that I'm interested in a, you know, office,  industrial, whatever it has, retail and so on.  And within that there is a certain amount of  workflow, right, in terms of identification of   that asset. Then you have, okay, I've identified  that asset. What do I think about the financing   of that asset? Or if you are a seller, how do I  think about the sale of that asset? Right? If you   are acquiring an asset, then you know there is  an aspect of managing that asset, both managing   as in physically managing the asset, but managing  the investment that you've made into the asset. 

And that's what I was talking about earlier. If  you recall, the three things that I said was,   okay, you know, asset identification and there's  documents around it. And as you think about those   documents, you know, what do you think about  how do you think about the workflows where you   can create some efficiencies or productivity  enhancements, if you will, as you think about   property management, You know, how do you  think about, you know, enhancements there?  And then as you think about it as an investment,  what are the opportunities? Let me take an   example, if you will. So again, going back to  our Smart facility management solution, right,   which I talked about, has about we have about  1 billion square foot using the building and   operations data. Right. This is part of that  contextualizing, if you will, the workflow,  

building it using the building operations data. We looked at actions such as automated, automated   maintenance, right? And so that meant that we  could reduce the tech dispatch, if you will,   by about 25%. And the operating expenses that  were there for this automated maintenance,   we were able to reduce the energy and the  maintenance by as much as 20%. So again,   we look back and say, let's look at all this  workflows, if you will, and look at opportunities   where you could bring II into the mix and that can  actually create value for it for the organization   and create value for our clients and deliver much  better differentiated outcomes for our clients.  So on that point, we have an interesting  question from Lisbeth Shaw on Twitter who asks,   Can you provide some examples of how AI has  changed the way real estate services, the real   estate services and investment business works? as you look at some of these interventions,   Right, Because I mean, it's a great question,  but again, it spans across multiple aspects,   if you will. So, if I go back to what I said  right. Is you look at things such as documents.  And so, for example, we have today internally  created automated content creation capabilities,   right. That take us that used to take two weeks,  but now take 2 hours. You know, I talked about,  

you know, again earlier about taking this smart  facilities and and, you know, saving operating   expenses such as energy and maintenance  by as much as, you know, 20%, if you will.  But again, similarly, we had a health care  client where we utilized dynamic cleaning and   we saved 11% savings over the baseline that we  have. So, again, it's the opportunity to really   look at that data, look at that workflow and say,  where exactly do you bring into the middle here,   into as an intervention, as an opportunity, if you  will, to really create some workflow productivity   enhancements and, and some savings, if you will? Can you drill in a little into the mechanism? What   is it about these tools or the technology or  the or the way you're using the data that, for   example, enabled you to go from what did you say,  two months to two weeks or something like that?  I don't remember the exact numbers you said. About two weeks to 2 hours, right? So in a sense,   what you are doing is you're looking at the  workflow and you're looking at the existing   set of documents that you're looking in, in  essence some form of pattern, if you will,   to say, look, the more I can learn about  this and to a certain extent extend.  Michael, it's also about the scale. How do you  really think about it? So when you have a scale,  

then you can identify certain patterns in  those patterns, start giving you insights into,   you know, what is that workflow, repeatable  workflow that is happening here and how you   can how you can create that workflow or  operationalize or optimize that workflow.  And in essence, you have to have three things  from a foundational perspective that you have   to bring into the picture here. One  is the data capability. So, you know,   it doesn't happen overnight, right at CBRE,  our focus on data and having enterprise grade   technology has been there for a long time. Right.  Which has allowed us to move quickly, you know,   if you will, into AI, especially junior area like. So that data capabilities having an enterprise   data platform where we can transform  data at scale, understand it at scale,   get insights from this data scale as a key pillar  of our foundation, if you will. Right decks on.   The second thing is really looking at is setting  up a generally the AI platform and, you know, not   necessarily trying to solve a one-use case if you  will, but understanding, if you will, patterns. 

I mean, we've launched industry's first large  language model, if you will, multi-language   model interface, right? And we've got we focus on  rather than focusing on on one specific use case,   we've made it self-service internally so that  people could utilize it. It has many of the   same features as CBT. We made it in such  a way that people you could utilize it for   many different use cases, if you will, right? And then the third piece is how do you really   go back and look at utilizing the information,  the insights that you got from those deep data   capabilities to say in this workflow, what is  the end result that needs to be there, right?   And what does the end result really mean for  the customer that is utilizing that end result?  So I think across those three  things, the data capabilities,   some form of analytical A.I. to really think  about the usage and then the Jenny AI platform,   if you will, you start looking at that pattern  recognition, you start looking at that workflow,   and it's not a one size fits all. And,  you know, situations are different,  

but, you know, once you have a certain  approach to a certain pattern to it,   it starts with giving you that value, if you will. We have a really interesting question from   Arsalan Khan on Twitter. And Arsalan is a  regular listener and he always asks very,   very thoughtful questions. You're working with  such a large body of data. How do you ensure that   you have confidence in the data? Because you're  ultimately going to be making potentially very   large decisions on the basis of that data. We have, you know, a lot of internal data  

that we would triangulate with where  we have data Partnerships are real,   if you will. Right. And those data partnerships,  what data we have, what data can we use,   and third party data, if you will, data that we  gather through partnerships allows us the ability   to be able to triangulate some of that data and  understand the call it the integrity of the data.  And if you go back, if you will, right where I  said was look, as in when you set up a a culture,   not every use case will be successful in  the beginning. Right. And in some cases,   we've actually had to we've learned fast that  there are certain type of data, if you will,   that either isn't collecting, collected at the  level of granularity that is required for this   use case or isn't being collected at the level  of comprehensiveness, if you will, in in this.  And then we step back and said, okay, what are  the you know, what are the proxies, if you will,   for that kind of dataset that we can use  instead of, you know, the scope that we   were originally thinking about? Could we think  about it in a smaller scope, if you will? Right.   And so that's where we've stepped back and  said, most important in our focus has been,   you know, our focus previously and continues to  be around as one key aspect of our data platform. 

This is again, I stressed back on the enterprise  data platform that we have is we have to step back   and say, can we enable this unique use case  through the data that we have access to? Are   there opportunities for proxies? Are there  opportunities that we need to fundamentally   think about the workflow, not just in terms of  the workflow, the existing workflow, but do we   need to change the workflow that we have in order  to be able to collect this data going forward?  Or are there proxies, if you will, that can give  us an opportunity to make some advances in the   use case that we have? So but a great question.  You know, if that's a very key, I would say path   forward in focus, if you will, that we have in  terms of how do we really think about our data,   both what we have, what we can use and that which  we can source, if you will, to our partnerships?  So, the partnerships aspect is also another very  important foundation of your ability to execute   on this kind of a AI data driven strategy. Absolutely. Absolutely. Michael. I mean,   we don't believe that, you know, we just  have to go it alone. Right? It's it's we   have a very strong build by partner strategy,  right? I mean, we don't believe that. Again,   we have to do it. All right. At the end  of the day, it's about the time to value  

and the client outcomes that we really focus on. And to that extent, you know, again, going back   to that strategic alignment that I talked about is  we look at those strategic CBRE priorities and we   say in order to achieve these, in order to get  to these strategic priorities, what do we have   as capability today? And who are the potential  partners that we need to have in the marketplace?  And we scan the market on a regular basis  for partners as well as emerging companies,   if you will. Alison Bell, who I work very  closely with, you know, is a global head   of our digital strategy, tech acceleration  and partnerships. And Alison and her team,   you know, regularly look at the market from,  you know, from our digital strategy perspective   for what we need in terms of partnerships. And today I will say that, you know, we have  

partnerships with all the major tech providers, if  you will, and it has given us an edge, you will,   in terms of early access to capabilities such as  large language models or other gen AI capabilities   that we have integrated into our ecosystem of  the digital and technology stack that we have.  I think what we've also made to share, we've  also made selective investments, if you will. So   we made a $100 million investment and invites. We  were in the leasing and property space as well as,   you know, partnership with Key, if you will.  Deep is a company that collects energy,  

waste and water data. So if you think from a  sustainability perspective, that's very important.  So again, to bring it back to your point,  again, it wasn't it isn't about, you know,   building everything internally, if you will.  It's about the time to value to the market.  The DNA of CBRE, of course, is about commercial  real estate, and you're describing ways of   working and ways of thinking about this data,  this data asset that is very different from   the skill set and the background and  the culture of commercial real estate.  So how do you make that kind of  transition where the folks in   the company can learn to think this way and  learn to think about data as being an asset?  Yes, it's relatively new technology. But if  you think about other previous technologies,   Right. I mean, digital transformation, right.  About utilizing data at scale, if you will. 

Or about being able to capture other data in a  large scale data at scale, right. I mean, at CBRE,   you know, we're all about physical assets, And physical assets have a lot of data associated   with them. Right. The location, you know, location  being the address, if you will, or the floor or   within the floor of what exactly, you know, if  you have more than one, you know, company on a   given floor, you know, what exactly is part of  one versus the other or previously even Iot.  So, I think it ultimately goes back, if you will,  to say, hey, what exactly are those strategic   priorities? And talking about, you know, how you  think about AI as a potential solution. So again,  

we need to step back and this is something that  we have focused very much on, is stepping back   and really defining the problems statement. And so once we define that problem statement,   that's when we bring it into the mix and  say, is the AI the right solution here?   Is is a different technology the right solution  here? Right. We don't have to, in essence, bring   it to bear on each and every question, if you  will. Right. In some cases, for example. Right.  I mean, if we're trying to build a simple  dashboard that gives people, you know,   certain metrics. Right. Simple, quick metrics.  Right. And the first, you know, you just want   to be able to quickly build a dashboard and  then you say, okay, how do I make it better,   Right? How do I make interaction with this  dashboard better? That's where you utilize   a chat bot like capability to be able to,  you know, query the data in the dashboard.  Again, it's, you know, I would say there's a bit  of maybe a bit of a misunderstanding out there   that, you know, commercial real estate, you, you  know, isn't as far advanced, if you will. I think  

it's all about, you know, our applicability,  if you will, of digital transformation or AI,   if you will, to the solution. The one other  thing you kind of touch based on, and it's a   slightly bit of a digression, is, you know, there  is this aspect that every company needs to be a   tech company and I have a bit of a fundamental  call it a personal disagreement with that.  Which is to say it, not every team needs to be  called a tech company, right? Not every company   needs to be called a tech company. Right. It's  about how do you really utilize tech in order to   operate within the workflows, in order to operate  within the industry that you are working on?   Right. We are a commercial real estate company. We bring technology to bear. We bring it to bear   to really help our clients achieve certain  outcomes for us to achieve certain strategic   business priorities. Right? And that's one of the  many tools that we have at our disposal to be able  

to achieve the outcomes that we need to achieve. think you're being very clear that the fundamental   strategic goals of the company have not  changed because a guy has been introduced   into the mix. And so everything you're  doing with these tools continue to support   the ongoing strategic goals of the enterprise. Exactly of our customers. At the end of the day,   it's about, you know, I know I've  said strategic business alignment. 

Other part of that, I said client outcomes.  At the end of the day, it's about how do we   serve our customers better that customer  centricity, right? And if you do a great   job at being customer centric, if you do a great  job at really focusing on solving the problems   that your customers are really facing, thinking  about or need to plan for, then I think you set   yourself up in the right for being successful. Arsalan Khan comes back and says, How do you   know? Or figure out that the recommendations  made by the AI based on the data are correct   or not? And he's asking if you have any kind  of a governance process in place for this,   especially given the fact that you're working with  so many different partners and you're sourcing   data from different organizations. One is about data and the other is   about validation of the outcomes rate of the  recommendations, right? So one is about data   and the other is about, you know, the model  that you appointed to that data. You know,  

is that model really giving you the  answers or is serving you appropriately?  In terms of the interest rate? I think  from a data perspective, as I said, right,   we have the benefit of having a large  treasure trove of data, if you will,   of 39 billion points. So where we are, you know,  as the largest commercial real estate company and   a bit of a unique situation, if you will, to to  triangulate, to validate some of that data and be   able to utilize both internal data that we can  use as well as the external data partnerships.  So, there's that validation, if you will, from  that perspective. I think the second part is   about model and models are iterative,  right? So you look at and say, look,   that model recommendation that I got right,  you, you iteratively build a model, if you will,   from a small scale to operating at a larger scale.  And this is what I was saying earlier, right?  It's easy to operate at a small scale, but  it's complex and to operate at a large scale.  

And so from that model perspective, you  really, really look at iterative approach,   if you will. So, you say, look, you know, let  me apply this particular model in one particular   building, if you will, with one particular  client, if you will, and see what's results.  Do I start to get right Because inevitably  what you will find is and this, by the way,   is not specific to commercial real  estate. This is very broadly What   you do is when you're building a model, you  will find that there are certain aspects,   certain factors that you did not take into  account or you did not know to take into account.  And that comes when you're trying to  operationalize the model, right? Because it's   one thing to build a model and then it's another  thing to actually put the model in practice and   operationalize it. And so from that perspective,  we, you know, we have constant feedback loops,  

constant validation loops. And this  is where I was saying earlier also,   especially in the context of generative the AI,  you have to be careful and you have to have that   validation and you have to have that human in the  loop, if you will, to make sure that you know,   you're not getting in some form of hallucination. I know I'm digressing a little bit from a model,   if you will, to do generative AI, if you will.  Right. But again, this is about creating the   right feedback loop, making sure you understand  the right business problem that you were trying   to solve with the model, making sure that you  have the right data points in order to assess   the right business problems and to assess  that the model is actually performing on.  I'll give you an example, which I can't talk much  in detail about, but at a high level, you know,   we were working on we've been working  on this particular model, if you will,   for one of our businesses. And the question  one of the feedback that we recently got was,  

yeah, you know, we can look at documents  from this resolution or this perspective.  What if you change the document orientation, What  happens? And again, this is exactly we already in   that operationalizing the model really comes into  place, right? So hopefully I've asked Arsalan,   I've answered Arsalan’s question. You know, I  think Arsalan, you should definitely connect   with me on that because I'd love to learn  more. And I think anybody who has a great   thought process, a very clear, structured  thought process, I'd love to connect with.  We have another question from Twitter, and this  is you've been talking about the relationship   between the company's business strategy and  your A.I. strategy, but how do they influence  

each other, your business strategy and the AI  investment strategy? How do they work together?  You're not out there to say today I'm  going to have a use case right? I mean,   because if you go out there and see IVR  use case, then you are simply going to look   for a place where you can go invest it. But you need to step back and say, well,   what is the business strategy? Where can  I go solve that business strategy better,   faster, in a larger scale? How can I create that  impact, you know, for faster time to value? How   can I be more cost effective? I drive higher  productivity or can I bring new solutions,   if you will, to bear because they were  previously very complex, if you will.  And so in from that perspective, you say in  order to really achieve that business strategy,   diverse set of tech investments and  as part of the tech investments,   these are the set of investments I need to make  to be able to achieve those three things and   faster time to value ability to deal with larger  scale and be able to solve more complex problems.  You know, there is this blog post and that this  reminds me about Bill Gates has those Gates notes   that subscribe to and they land in my email. You  know, every whenever she publishes in one blog   post from about a year ago that he wrote and he  said, you call it the age of A.I. has begun. Very   interesting. It's it's a very comprehensive one. I would suggest people to look, you will find  

that age of AI has begun and read it is this  you know, he talks about among other things,   he talks about the fact that, you know,  the use of AI is especially relevant in   systems that have scale and complexity that is  tough and hard for humans to really deal with.  And he talks about the example of he talked  about the example of drug discovery. Right.   Or health care. Right. In which you have complex  biological systems. Right. That are very tough for  

us as humans to be able to do, be able to track  and comprehend on that level. Right. So, I think,   again, I go back to the scale and complexity. And in that context, second thing is   personalization is what he talked about, right,  especially within the context of the societal   good about education. Right? How do we really  tailor content? How do we how do we track,   you know, how much engagement a student  has with a certain content and how do   we track it so we can tailor content for  better educational outcomes, if you will. 

How do you think about the ROI of your  investments in AI? Is there anything are there   any distinctions how you evaluate air ROI relative  to other investments that the company might make?  There's one aspect, culture of experimentation,  right? Learning fast sometimes, you know,   that precision that that you can have with certain  business cases to be able to say, look, if I were   to go down this path, this would be precisely the  impact I could make and this would be precisely   the outcome that I could get and therefore  precisely that return I could get right with.  As with any new technology and with AI, you  may not be able to get that level of precise   answers about what the impact of that scope  will be. And so I think fundamentally, no,   there is no difference, right? When you think  about it, the fundamentals and what I mean by   the fundamentals, what you are tracking  to make sure that you look at the ROI,   which is things like, you know, productivity or,  you know, top line growth or things like free cash   flow that can be generated or client outcomes  like better client experience, if you will. 

Fundamentally, those metrics should  not change. Right? And we don't think   about it. You know, they must be different. But the aspect is that if you look at 180 use   cases, you may be able to bring that precision to  a certain number of them in and in other cases,   you just have to go forward and say, look, you  know, this is about experimentation, this is   about learning faster and this is, you know, if we  don't go through this door, we're not going to be   able to understand what lies on the other side. And so hopefully that answers your question.  

Michael. It's yes, fundamentally the same.  But keeping in mind that when faced with a   new technology that you have to make, you have  to take certain you may not have the most precise   answering part of the use case and if you will. So, part of that is the cultural change responding   to the nature of the technology. Yes. Exactly. Exactly.  My advice for anybody would be to say, look,  you know, as you look at the opportunity for   it to drive impact, especially transforming,  you know, the way in which we work or in   terms of helping us gathering insights, driving  predictions, we just have to step back and say,   no, we need to we need to have a culture of  learning fast. We need to have the culture of  

moving and pace and urgency. And sometimes that  may mean that you seed a separate investment for   these kind of things so that you can do that. And  equally importantly, something that we've done as   CBRE for a long time is, you know, making sure  that you invest in the right data in enterprise   grade technology stack and then finally is  making sure that you continue the context,   realize your problems and your solutions. You contextualize your solutions to the   use cases, to the industry that you  are in and the workflows that you. 

Being clear about the outcomes, the  processes and the workflows, as you said.  Exactly, exactly. And you know what? It sounds  easy, but it's not like because sometimes,   you know, there's this, hey, let's go shade change  the shiny object, if you will. And that's what  

I'm saying is let's not chase the shiny object. Let's figure out where that application is and   what is the value of that application. Let's  continue to move forward, but be mindful,   be pragmatic, be strategic about it. Actually, it does not sound so easy   to me because you have an existing set of  clients, set of processes and revenue drivers,   and the company works in a particular way. Now  you're talking about introducing a potentially   significant change which is disruptive, but at the  same time you don't want to disrupt what works.  Yes. And that goes back again to what I was saying  about operationalizing. Right. I right. You have  

to think about in the context of how do you really  bring the the technology you bring, if you will,   bear in that operational context And what are  some of the aspects you need to be mindful about,   you know, when you try to operationalize it? So, it is this is again, why it goes back   into you're looking at what are your strategic  business priorities and what are the workflows,   if you will, right, that you have? Right.  I mean, you know, in my prior lives,   I've served as a chief product officer and I, I,  I often talked about that. You know, I want to   take the work product management and change it to  workflow management because at the end of the day,   you're building products for a certain workflow. You're trying to solve for a certain improving   sort of workflow. So should we call it  workflow management with digital technology   and or is it digital and technical digital  applications and write the best technical   technology enterprise grade technology Stack  Yeah, maybe part of that is that right?  We have one final question from  Twitter again, from Arsalan Khan,   who again asks a very thought-provoking question.  And it looks like he's going to have the final   word here or the final question. What happens  when an executive uses their vote to not do  

anything with AI inside the organization?  How can non-executives change that veto?  You have to step back. It's first and  foremost understand why they did that, right?  Because if you're trying to solve simply as a  knee jerk reaction and say or look, you know,   somebody said such and such thing and therefore  I need to have an immediate reaction because,   you know, my belief is that the problem  requires a different solution. Then you're   going to find yourself in a tough spot.  I think the first and foremost question,   the first and foremost approach is to understand  why that veto happened in the first place.  You know, there might be some legitimate  reasons, right? You know, again, as a you know,   I always coach and advise, you know,  people is to say, first and foremost,   understand the industry, understand the  workflows, get into the details of the workflows,   if you will. Right. And so is it the fact of,  hey, that veto happened because, you know,   the solution that is being designed  is is going to is going to be great,   but it doesn't have the right level of  operational considerations to it for deployment. 

And if that's the case, then you go solve  that problem. If the problem is, hey, yeah,   the solution is great, but it only solves  part of my problem. And in order to, you know,   you're asking me to fundamentally change 100% of  my workflow while you're really going to bring   a solution that only solves 10% or 20%. And I'm  not going to go disrupt people who, you know, are   used to a certain workflow only to be disrupted  for 10 to 20% of the workflow, then that's it.  Then you have to go answer that question  differently. So, I'd say the first and  

foremost thing is the art of asking questions,  right? Step back, ask the question, understand   the reason behind the veto, right, if you will,  because nobody ever said that, you know, hey,   I don't want to do things. I don't want to grow my  revenue, or I don't want to have a higher margin   or I don't want to have higher free cash flow. There's nobody out there who says that, right? So,   there must be a reason behind it. And I  think it's imperative, first and foremost,   to understand that reason. Lot of wise advice right there. 

And with that, I want to say a huge thank you to  Satnam Singh from CBRE. Satnam, thank you so much   for being here with us today. Thank you for having me,   Michael. It has been a pleasure. Everybody watching, thank you for being   such great audience and to the people asking  these excellent questions. Thank you so much.   Before you go, Please subscribe to the CXOTalk  newsletter. Subscribe to our YouTube channel,  

leave comments and check out We have  incredible shows coming up. Thank you so much,   everybody. I hope you have a great day  and we will see you again next time.

2024-02-26 07:28

Show Video

Other news