WEBVTT NOTE Sentence-level transcript of https://www.youtube.com/watch?v=Fe4aSMXsOZ4 NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/Fe4aSMXsOZ4.html. A cue ends where the next begins, or 2 s after its last word. s1 00:00:01.480 --> 00:00:02.040 Hello. s2 00:00:02.040 --> 00:00:07.040 I'm very excited to give this talk today at AI Engineer NYC 2026. s3 00:00:07.040 --> 00:00:10.720 Um this is a collaboration between Sprana and Neo4j. s4 00:00:10.720 --> 00:00:12.640 Sprana, I am the founder of. s5 00:00:12.640 --> 00:00:14.640 It was founded in June 2026. s6 00:00:14.640 --> 00:00:19.120 Neo4j is a graph database in based in Sweden, Malmö. s7 00:00:19.120 --> 00:00:27.440 And um a special thanks here at Zaeem, who is the collaborator on Neo4j's side. s8 00:00:27.440 --> 00:00:34.000 This is about every building is a balance sheet and the spatial memory for real-world assets. s9 00:00:34.000 --> 00:00:40.440 The largest asset class on Earth is real estate and it's about 400 trillion globally in size. s10 00:00:40.440 --> 00:00:45.160 And um this is more than global equities, debt, and gold combined. s11 00:00:45.160 --> 00:00:48.080 If we look on the right, you see a lot of squares. s12 00:00:48.080 --> 00:00:54.680 Each square is about 1 trillion uh in worth and you can see the residential part is the biggest. s13 00:00:54.680 --> 00:00:56.520 It's around 300 trillion. s14 00:00:56.520 --> 00:01:02.160 Then you can see the commercial is around 60 trillion and then agriculture around 50 trillion. s15 00:01:02.160 --> 00:01:11.440 And uh today we will focus more on the commercial side, but all of this um real estate is impacted by essentially a gap. s16 00:01:11.440 --> 00:01:26.840 And the gap is that our financial books uh where we have lending and insurances and um mortgages, all of this is book kept um and this is updated every quarter. s17 00:01:26.840 --> 00:01:34.200 But the actual underlying assets, which is the real estate that they are based on, um they're not updated every quarter. s18 00:01:34.200 --> 00:01:45.400 In fact, they're only updated maybe every few years, maybe just before the financial instrument for the for the real estate is being created. s19 00:01:45.400 --> 00:01:58.560 And also maybe again once it needs a modification or a new financial instrument on top of it, maybe in in year five is a very re- realistic time frame. s20 00:01:58.560 --> 00:02:07.960 But between year zero and year five, you really don't have an update on the building state or the real estate state. s21 00:02:07.960 --> 00:02:17.680 So that means that while the book closes so many times, there's really a black box or maybe something that is based on some pictures and some PDFs about the building. s22 00:02:17.680 --> 00:02:19.959 And this is a real gap. s23 00:02:19.959 --> 00:02:22.400 Um and you can see it. s24 00:02:22.400 --> 00:02:29.400 So according to Kroll, um they did a research with 1,500 properties. s25 00:02:29.400 --> 00:02:32.840 Essentially, 70% of the buildings are underinsured. s26 00:02:32.840 --> 00:02:37.520 And why exactly that is, um I don't exactly know. s27 00:02:37.520 --> 00:02:38.560 It didn't conclude. s28 00:02:38.560 --> 00:02:44.560 But my read is that the the record is quite old and it's maybe also not as detailed. s29 00:02:44.560 --> 00:02:51.240 So the insurance is not very highly specified and really fulfilling the needs of the real estate. s30 00:02:51.240 --> 00:02:51.611 And s31 00:02:51.611 --> 00:02:51.760 [snorts] s32 00:02:51.760 --> 00:02:56.080 this impacts around 4 trillion US commercial mortgages. s33 00:02:56.080 --> 00:03:00.320 And um yeah, every one of them rests on the building. s34 00:03:00.320 --> 00:03:05.400 Uh so the question, what would it take to make a building as queryable as a ledger? s35 00:03:05.400 --> 00:03:09.400 And a ledger here can be give you an answer in milliseconds. s36 00:03:09.400 --> 00:03:14.120 But a building today is giving you only an answer after a site visit. s37 00:03:14.120 --> 00:03:23.000 And even then, maybe in like some notes, PDFs, very dispersed and intransparent, not unified. s38 00:03:23.000 --> 00:03:26.720 Um so this is where the idea comes in, where every walk becomes a memory. s39 00:03:26.720 --> 00:03:32.800 And how this works is you just take a off-the-shelf sensor, you walk the space, and you map it. s40 00:03:32.800 --> 00:03:35.239 And you can see it up there actually being created. s41 00:03:35.239 --> 00:03:38.320 That was me uh going through a conference hall. s42 00:03:38.320 --> 00:03:42.600 And the whole walk or the whole scan here took literally just walking through it. s43 00:03:42.600 --> 00:03:45.080 And a minute later, you had the scan. s44 00:03:45.080 --> 00:03:51.080 Um so you walk through the property, whoever that is, property management, etc., can be equipped with it. s45 00:03:51.080 --> 00:03:52.320 They map it. s46 00:03:52.320 --> 00:04:01.040 And [clears throat] as they map it, you get all the assets in the building being recognized and stored in a database, which is building below. s47 00:04:01.040 --> 00:04:03.280 This is the graph database. s48 00:04:03.280 --> 00:04:09.520 And so, this is then has a semantic meaning, which means we can really understand the space. s49 00:04:09.520 --> 00:04:12.840 And as we walk it again, both walks are kept. s50 00:04:12.840 --> 00:04:16.959 So, walk on Monday and a Monday next year or the next quarter. s51 00:04:16.959 --> 00:04:19.239 And these are now aligned on a time axis. s52 00:04:19.239 --> 00:04:29.919 So, we have a space, we know what's in the space, and we know the evolution of the space according to every every time we walk it and we measure it. s53 00:04:29.919 --> 00:04:31.560 And this is all kept as a graph. s54 00:04:31.560 --> 00:04:40.320 Graph have a bunch of technical benefits, but one of them is they're extremely flexible in what information you want to gather and how you want to connect it. s55 00:04:40.320 --> 00:04:47.320 And so, this can be hyper personal personalized to each asset to each to each property. s56 00:04:47.320 --> 00:04:48.840 Um yeah. s57 00:04:48.840 --> 00:04:53.720 And so, now what is it that a record has to answer? s58 00:04:53.720 --> 00:04:57.840 So, these are not vision questions, they are ledger questions. s59 00:04:57.840 --> 00:05:02.919 You really want to know in a property or in the real estate, what has changed? s60 00:05:02.919 --> 00:05:09.000 So, two walks of one space compare object by object and and surrounding by surrounding. s61 00:05:09.000 --> 00:05:10.720 You also want to know when it has changed. s62 00:05:10.720 --> 00:05:14.440 So, every observation carries and captures where it came from. s63 00:05:14.440 --> 00:05:21.360 And you also want to know who needs to know information or deltas that have happened that we have now observed. s64 00:05:21.360 --> 00:05:41.560 So, when you find something with evidence, for example, structural damages or maybe for fire insurance that fire extinguishers are off off Yeah, they have expired, then you need to know that and you need to be informed and counteract on it. s65 00:05:41.560 --> 00:05:49.920 Or maybe the insurance can know and they can increase their premiums by the performance of the upkeep of the property. s66 00:05:49.920 --> 00:05:53.120 So, hyper-individualized insurance. s67 00:05:53.120 --> 00:05:55.040 Um and here's really how it looks. s68 00:05:55.040 --> 00:06:01.280 So, this is just me walking through a kitchen captured twice, once on the left and once on the right. s69 00:06:01.280 --> 00:06:12.160 And you can see you just walk the space, you directly get the map, and um and as you walk the same space, they're being recognized as the same space, s70 00:06:12.160 --> 00:06:20.680 and you can align them between um the same space in the time period A and the time period B, and you have a timeline. s71 00:06:20.840 --> 00:06:31.440 And once you have a map, you can now use it as an AI-ready asset, which means that you can chat with it, you can talk with it, you can ask questions against it. s72 00:06:31.440 --> 00:06:34.400 For example, here I ask, "Where can I heat myself?" s73 00:06:34.400 --> 00:06:37.960 And in the map, the answer is being given to the radiator. s74 00:06:37.960 --> 00:06:41.200 Or, "Is there a glass or mug to drink from in a kitchen?" s75 00:06:41.200 --> 00:06:45.280 It It points you to the glasses and mugs in the kitchen. s76 00:06:45.280 --> 00:06:53.160 Um so, this is very powerful because it makes the map from pixels into something we can work with. s77 00:06:53.160 --> 00:06:55.800 Um and now also the change detection. s78 00:06:55.800 --> 00:07:03.960 So, this were two passes we made of this kitchen or of this space, and we can really now connect these two and ask, "What is the delta between them? s79 00:07:03.960 --> 00:07:05.919 What has changed in this environment?" s80 00:07:05.919 --> 00:07:24.720 And in this case, um these three chairs in the middle have been displaced, and you can see when you chat with the with the asset, with your data, you can actually identify the delta and get the ground source reference to the actual capture, so you have proof of it changing. s81 00:07:24.800 --> 00:07:26.560 Um so, what does this all mean? s82 00:07:26.560 --> 00:07:31.960 It means that you have now a building that is an AI-ready data asset. s83 00:07:31.960 --> 00:07:37.800 So, there's one record of what it what is there, and you can connect it to any downstream task. s84 00:07:37.800 --> 00:07:47.640 So, on the left we have this conference scan that I made, and then you can now check over time any repair that needs to be happening or some maintenance s85 00:07:47.640 --> 00:07:59.360 or some compliance checks or insurance or maybe interior design changes or maybe even you have some robotics that you want to place there and it's like a digital twin for it to use. s86 00:07:59.360 --> 00:08:02.160 So, let's check the insurance. s87 00:08:02.160 --> 00:08:14.040 When you walk the building, when the policy starts, you have like a first initial state of the property and on based on this you can do a hyper-personalized insurance policy. s88 00:08:14.040 --> 00:08:23.400 And as you walk it again at renewal, you now see the delta and the upkeep and so on and the insurance policy insurance company can actually measure the performance s89 00:08:23.400 --> 00:08:39.680 of the property for relevant facts against its policies and give a very hyper-personalized insurance policy, which means a very efficient premium to insure such property and insure it in the right way. s90 00:08:40.159 --> 00:08:42.400 Uh yeah, and these are the numbers behind that. s91 00:08:42.400 --> 00:08:58.160 So, the business case here for the insurance for example would be that the insurance pays for what it can verify and so currently what insurers know is that one in three commercial buildings are on the books with the wrong construction class. s92 00:08:58.160 --> 00:09:10.520 35% of the buildings in the specific case for example for the hurricane premium can actually get a deduction of of such percentage if it's a certified building. s93 00:09:10.520 --> 00:09:20.080 Certified building means it has certain compliance standards and that you can check from such an AI digital asset of the of the asset. s94 00:09:20.080 --> 00:09:35.080 And you also see that 70% 73% of the insurers agree that they claim fewer claims for hurricane damage if they have been checked and certified against the building. s95 00:09:35.080 --> 00:09:48.040 And just this is just one such certification you can do automatically with AI plus a lot of other certification and inside of the building that qualifies maybe for such programs. s96 00:09:48.040 --> 00:09:55.120 But yeah, all three of them essentially depend on somebody knowing the building and and the state of the building. s97 00:09:55.120 --> 00:09:56.720 Um and that's it. s98 00:09:56.720 --> 00:09:58.080 That's essentially the talk. s99 00:09:58.080 --> 00:10:11.560 I think that the real estate class is very big and it really needs to be transparent and there's a huge inefficiency happening day that I think with the new insights of what AI is capable to do s100 00:10:11.560 --> 00:10:20.480 we can make much more transparent and create a really new hyper-customized market around around assets. s101 00:10:20.480 --> 00:10:22.000 Thank you very much. s102 00:10:22.000 --> 00:10:29.160 Um if you want to find out more, please visit my website spranik.com and thank you for your attention.