 So I think we have a critical mass now, so I'm going to go ahead and get started. As I said before, my name is Felicia Polum from CBP, and we're moving into the industry presentation phase of our day. So we'll hear from each presenter for 15 minutes, and then have a 10-minute opportunity for questions. So the two standing microphones right there in the aisles are to ask the questions, so please feel free to get up and get in a queue to start asking your questions when the time comes. But we'll also have someone walking around with a microphone in case you're unable to come to the standing mics. If we run out of time, I would encourage you to please see our events support team, someone wearing a green lanyard in the rotunda to ask your question so that we can follow up and make sure you get an answer. So then finally, as we move into the presentations, I want to share a brief disclaimer. We have a much longer disclaimer here on the screens, but in short, the purpose of this event is to provide information and a platform for sharing across the private sector. CBP is not endorsing any particular technology or technology provider. So thank you again for joining us today. We're really excited and appreciative that you chose to spend the day with us. And now I'm going to introduce our first presenter. So I'd like to introduce Mr. Kitt Conklin. He's a Vice President at the Research and Data Analytics firm, CARON. He leads engagement with CARON's government clients and works with global corporations and financial institutions on forced labor, sanctions, and supply chain risk. Kitt is also a senior fellow with the Atlantic Council's Geotechnology Center. Prior to CARON, Kitt served in multiple national security positions with the U.S. government. He has lived in Beijing and worked on China for more than 15 years. Thank you. Good morning, everyone. It's a pleasure to be here. My name is Kitt Conklin. My background is in the intelligence community, and today I'll be discussing CARON. You can think about us as a research and data company, certainly, but in reality, we're an intelligence company. We specialize in detecting risks within global supply chains. And I think before we show any of the technology itself, it's useful to start a bit more about our analytical trade craft. How do we detect forced labor? What is the intelligence process that we utilize to create this data? And then I'll discuss a bit more about how we work with a lot of the partners and companies actually in this room. We are a complement to a lot of the supply chain mapping companies that are out there, and we can answer any questions that you may have on that. But to start, let me, I think, contextualize a bit about the UFLPA in general. We had some great presentations this morning that discussed the background on the law. How do we get here? What is the rebuttable presumption? Now, I want to discuss where that law and due diligence intersect. So forced labor networks, when we think about detecting these types of risks within global supply chains, they don't just exist randomly only in Xinjiang. It's global in nature, right? There is no de minimis requirement or threshold for a UFLPA detention. So if you've got 0.1% of your commodity that contains elements or inputs that were made with forced labor, you could be subject to a detention. So how do you actually detect that risk? A lot of people kind of forget, right, that CBP has been very prolific with respect to FAQs and due diligence best practices and how to actually detect this risk. So we at Caron use that government provided risk thresholds for our analysis. So we've got a team of researchers, dozens of analysts, they speak Chinese and they spend all day every day looking at open source information to map out the risks associated with forced labor typologies and indicators. We use primary source information to do this. So we don't have anybody in China. We don't have anybody looking at these issues inside of factories in Xinjiang and other provinces in China. Everything that we do from an intelligence perspective is using human analysts to find open source information about the risk typologies that CBP has published. So let me spend a bit of time talking about that. CBP I think gets hit hard sometimes by those in the industry for not telling us more about where the risk is. And I would disagree with that. I think fundamentally CBP and the Xinjiang supply chain business advisory originally published in 2020 that guidance was updated in 2021 and now the due diligence best practices have been incorporated both into the UFLPA and subsequent guidance from CBP. It's been very specific about what companies should look for within global supply chains that cause risk. So let's dive into these for just a few minutes. The typologies and warning signs here are from that government guidance. This is the roadmap for every UFLPA compliance program. You have to conduct that research. You have to identify where forced labor risk is per these typologies. So I mentioned that because we at Karan don't know where the risk is as the North Star, as the starting point for all of the work that we do when we're trying to detect this risk. So what does this mean? It means, for example, identifying companies that are co-located with prisons. So how do you do that? You first have to start by identifying all of the prisons, all of the internment camps, all of the re-education centers and then from there you identify what are those companies that are operating inside of that internment camp, inside of that re-education center. What are the companies that are located along the fence of that internment camp or that prison? Where are those laborers that are imprisoned in that facility going to work for 12 hours a day? And then once you identify those companies, it's a matter of following those supply chains, regardless of what the commodity is globally. So as you follow that risk, traditionally what will happen, if it's a raw material, if it's silica, if it's cotton, if it's anything else, lithium, aluminum, it doesn't matter. We'll follow all of those products as they enter the supply chain inside of China. So products manufactured in whole or in part, it may be a prison and then it's sold to another company in China. Other inputs are added. A widget is created. Raw material is turned into a whole product. And then maybe it's exported to Vietnam or Indonesia or Mexico. We follow that research wherever it takes us. The other starting points for our work are also outlined here. So companies co-located with prisons one typology, other things that CBP has regularly and consistently highlighted as being representative of risk, things like this program. So these are again starting points for our research. We're going to determine where are those companies that have received government subsidies to create new factories to manufacture or mine new products in Xinjiang and then from there that same typology following those supply chains. Same with other typologies. Frequently government and industry kind of thinks about the UFLPA risk from the perspective of only looking at inputs from Xinjiang. Well, that's not sophisticated. So we have to think about this from the perspective of the other warning signs that CBP has publicly highlighted. Labor transfers are the big one. So what this means is if you've got a company that is receiving labor transfers, ethnic Uyghurs, Kazakhs, Turks other ethnic minorities being involuntarily moved from their homes in Xinjiang and forced to work at other factories maybe in Guangzhou, Shanghai other provinces, other cities in China anything at that facility is now tainted with forced labor risk. So as you're thinking about due diligence and as you're thinking about structuring your program you need to ensure that you have intelligence on those actors located not just in Xinjiang but globally that are connected back into these typologies. From our perspective though at Caron when we provide our forced labor data for companies to screen against, to utilize and supply chain mapping solutions to include some of the great partners that we have here today the first thing that we're going to look at is understanding the ownership associated with that risk so imagine these red circles at the top of the screen maybe they're a company that's located at a prison or maybe they're a company that's received a labor transfer or maybe they're a company that's engaged in the cotton or silica trade the first thing that we're going to do is map out that ownership and then El Nagar earlier made a really great and interesting point that I want to highlight here we at Caron also follow these networks as they evolve so ownership changes regularly we see this all of the time front companies shell companies companies that have been added to the UFLP entity list or have been publicly associated by the great work that Sheffield does and Laura Murphy's team does once a company is publicly identified with these types of networks they don't just stop there's a vision activity that takes place so one element of our data that we provide is following these companies as ownership change as divestments change as front companies pop up obfuscating the source of that risk we'll also look upstream this is a very very common typology to understand and identify where for example companies are owned by the Xinjiang prisons administration for those that may not know the Xinjiang prisons administration as a government body they exert control along with the Xinjiang production and construction core over almost all of the prisons and reeducation centers in Xinjiang so why is that matter it matters because if you're looking upstream for a company in Shanghai it's possible that there will be exposure to the Xinjiang government officials that are responsible for prosecuting what the United Nations has described as a genocide in the country so you have to think about this risk physically the next piece that we at Keron will follow are control relationships so if you're thinking about the UFLPA only from a beneficial ownership perspective it's not as sophisticated as you could you could be doing it so another way that we at Keron map out this risk is by understanding for example intelligence that will never show up in corporate records looking for example at those companies and organizations that are owned or controlled by government officials understanding other risk typologies like for example operating units of the Xinjiang production and construction core these are not a publicly available pieces of intelligence that are only in corporate records you have to dive into very difficult to find sources of intelligence to map out these risks and then figure out from there where the source of that risk ties back into a typology that CBP has outlined another element of what we do and perhaps a bit beyond the scope of the discussion today is we take data on who is investing in these companies so if you're a US bank a US private equity company if you're in VC and you're looking to invest in startups or perhaps publicly traded firms in China we help folks navigate where that debt where those securities where those equities are that are tied back into the publicly traded entities in China that use forced labor because at the end of the day it's not just privately held companies that use forced labor thousands of companies in China are affiliated with UFLPA risk many of which are publicly traded but I want to spend a bit more time talking about the typologies that CBP has outlined that we at Keron use as our guiding point for the research that we conduct so prison supply chains I briefly discuss this but here's an example of put on remove the compliance hat remove the supply chain procurement hat remove that council hat remove the supply chain procurement that you're wearing and think about this from the perspective of an enforcement so how or why is a good detain at the border many in industry regularly ask CBP this question and CBP for all sorts of logical reasons is not going to be able to always provide publicly information about why a good has or hasn't been detained so here's an example just to make this a bit more real so if you've got for example so as a reminder that's a company that's co-located at a prison you have to understand as the source of that risk enters the global supply chain it's traditionally not coming into the United States or coming into Vietnam directly from Xinjiang it's going to intermediary companies tier 2s 3s from the risk itself so we at Keron map that risk out using again humans we don't have AI we don't use lots of big data we use humans that are intelligence analysts that speak Chinese to conduct this research so we'll follow the risks from the supply chain connected in a prison and it will follow that risk globally so earlier today CBP announced the dashboard if you look at the dashboard huge volumes of CBP detentions have been in Southeast Asia countries Malaysia, Indonesia, Vietnam, why, how it's because these products are made in prison supply chains and other typologies that CBP has publicly identified and then they're being exported to these third countries and then being shipped into the United States so if you're thinking about why a product from Vietnam is being attained it's because there could be some risk tied back into for example prison supply chains but there's other risk typologies and this is the one that we get asked about the most mutual pairing assistance this is not a term that Keran made up and it's not a term that CBP decided randomly to incorporate in their due diligence best practices for industry this is a term that's used by the Chinese government to help companies help companies in China receive government subsidies to manufacture and create new satellite factories in Xinjiang that then use forced labor to manufacture goods, components or widgets so what does that actually mean? It means for example maybe you have a publicly traded company maybe it's in Xinjiang maybe it's not in Xinjiang but they will receive government incentives to utilize ethnic laborers and then from that point you now have tainted supply chains that are again global in nature they don't necessarily need to only be in Xinjiang themselves the next typology that I'll spend a bit of time discussing is labor transfers another key typology that we have to spend a huge amount of time trying to identify if you Google companies that have been affiliated with labor transfers there's been some great work done highlighting these but there's a lot of information that is still out there that's able to be defined and we at Keran specialize in detecting these types of labor transfers so what does this mean it means for example you have a company located outside of Xinjiang or maybe inside of Xinjiang that has received an ethnic Uyghur ethnic Kazakh ethnic minority being forced to work in a factory like El Ngar and others discussed this morning in other provinces so now you've got the UFLPA risk beyond just things associated with Xinjiang zip codes and you've got to think about this from a strategic perspective when you're taking a risk-based approach to due diligence another key element that we at Keran detect risk again based on government guidance is associated with the Xinjiang production and construction core and this gets into the core of what I would describe as the unique nature of the UFLPA associated with sanctions risk in particular so the U.S. government through a variety of economic statecraft tools export controls sanctions the UFLPA the import bans WROs historically has used a broad set of tools to counter this type of activity the genocide that's happening in Xinjiang one element historically that was used first was sanctions for those that spend time in the sanctions arena there are very very different legal regulatory and compliance requirements for sanctions exposure than there is solely for UFLPA exposure interestingly when it comes to the XPC guidance that CBP has publicly pushed out though this is where the rubber meets the road sanctions issues and UFLPA compliance become one and many an industry that we talk to are very concerned about incidental exposure tied back into XPC affiliates next XPC subsidiaries there are thousands of XPC subsidiaries they operate hundreds of prisons and interment camps they provide power on behalf of the Xinjiang economy it is a massive paramilitary organization that should cause every compliance official that's concerned about these issues to raise their eyebrows it is very difficult to find intelligence on this it has taken us two years to use that human derived research to holistically understand every party globally that's connected back into XPC and so what this means for example is helping clients understand and navigate where is that strict liability sanctions exposure where is the risk tied back into XPC affiliates from the UFLPA perspective or broader regulatory and ESG risks XPC products are a big deal and it's not just cotton so for anybody here that's just thinking about cotton there are all sorts of other industry verticals that are relevant for XPC the kind of the final typology that I want to touch on is government subsidies so we at Keran will again start with that risk related information that CBP has published those typologies and warning signs one of which is government subsidies so we'll follow for example all of those companies that have received Xinjiang government subsidies for any product line a lot of people don't know or recognize that in Xinjiang if you are planning in China that you want to legally operate in certain industries it is almost required that you receive subsidies from the Xinjiang government cotton, silica, lithium aluminum there are others that we can discuss if you're interested you have to have government licenses by the Xinjiang government to operate in certain segments of these fields by mapping out all of those companies that have received those licenses you're able to get a very clear and concise intelligence picture on where the actual risk is for example for a detention and all of this information regardless of what typology it's connected for our researchers are global in nature so we don't just think about the UFLPA from a China perspective we have analysts that speak Korean, Vietnamese, other languages we start with the risk and then we follow that risk wherever it takes us so if you're concerned about your tier 1's, your tier 2's anywhere on earth there's a good chance if there's a UFLPA risk our analysts have looked into those networks and so we provide this data in a couple of different ways and this is my last slide here and I promise I'll end on time so how do you actually get the intelligence we work with nearly every company in this room we are not competitive with the companies here we're partners with the companies here we work with Greg and the Oratane crew or Grant excuse me and the Autonic crew we work with these folks we provide our data for integration into these types of platforms so Karan has a UI you could subscribe to our UI and search for a single party lookup and see if there's any risk for sanctions or forced labor but the power of the data is at scale and it's automated so if you have questions about our partners we've got so many in the room with us today Karan's data can be integrated into your existing supply chain mapping solutions your existing denied party screening solutions your existing internally developed solutions our risk data is agnostic and we're happy to provide any additional insights about who we are or what we do but let me pause there any questions I'm sorry I missed your presentation the first part of it but I've been listening in remotely Virginia from Miller and Chevalier I've met Kip before and I know that a lot of companies really rely on Karan data and use a lot I'm just wondering if you guys have incorporated information from secondary sources like the Laura Murphy reports so if there's an NGO report out there companies are using Karan can they rely on it to be able to tell them about that secondary reporting thanks so first and foremost I like everyone here should always read Laura's reports she will be going into that tomorrow but Laura's reporting is fantastic it's top notch we at Karan rely on primary source information so what that means for example is if there is a media article if there is an NGO report we will obviously read that because it's in our realm as an intelligence company to monitor those types of risks or those types of reports as an intelligence organization though we are grounded in the trade craft that we utilize inside of the intelligence community to map out this risk so I highlight that because our methodology starts with primary source information so our information is looking at Xinjiang government documents looking at trade information understanding inside of primary source information which companies have received labor transfers which companies have received subsidies from the Xinjiang government to work in the cotton industry or maybe operate a mine for lithium in Xinjiang we will look at that primary source information and we will rely on that primary source information to guide our analysis that said we like everyone here should be reading Laura's reports as soon as they come out they're fantastic any other questions I think they're awesome thank you for your time sorry one more my name is Omeed I'm with Z2 Data I believe we actually partnered with yourselves too so thank you for that you mentioned enforcement and this if you can answer this that would be great but if not you know who does the kind of onus fall on if I'm let's say an aerospace company who has association a supplier that's associated with the Xinjiang province who is that going to kind of fall on when it comes on enforcement are they working with me are they working with the semiconductor company yeah that would be great to know thank you I won't speak on behalf of CBP there's a lot of good people in the room that can answer that question a lot more strategically than I can but what I will say is that the onus is on the importer the onus is on whoever provides those inputs into that that widget that used force labor originally so there could be all sorts of regulatory and legal requirements and ramifications for a product that's detained at the border that's got any percentage of inputs made with force labor in it right so the folks at Miller and Chevalier and John Flitz organization there's a whole bunch of good attorneys in the room as well that can answer that question in more strategic detail than I can but what I will say and then what I would footstomp is that there's no de minimis threshold so it doesn't matter if it's 0.001% of a widget or a product that's being imported in the United States is derived from force labor CBP has the authority and the mandate to block that good or that product and so therefore it's on industry at all levels of the supply chain to ensure that the components and the inputs that are integrated into that final product do not contain those specific inputs in force labor and how individual companies structure that compliance program I've seen the full gamut but I would just footstomp again there is no de minimis threshold here for inputs made with force labor they all represent risk alright thank you all so much thank you again alright I'm ready to introduce the next speaker we have Mr. Grant Cochran he's a CEO of Ortein a company that's on a mission to be the global leading developer of Ortein and it's a company that's on a mission to verify of authenticity through science and data science Grant has used his diverse background extensive strategic knowledge and governance experience to oversee the commercialization of Ortein's scientific traceability service and the growth of the business globally for the past 12 years good afternoon thank you very much for the very kind introduction so today I'd like to go through about Ortein I'd like to start off to talk a little bit about our company who we are, where we've come from and what we're doing then I'd like to talk about how we add value to clients a little bit about the product and the science and then finally walk through a case study of how somebody uses our technology okay so what we've done is we have commercialized forensic science so the science came out of the criminal forensic field where it was used to prove where drugs come from, where bodies come from etc so all we've done is taken that well established science and applied it into pharmaceuticals food products and fashion products the company was formed in 2008 in New Zealand by Professor Fru Professor Fru is a leading scientist and the use of stable isotopes and other techniques previously worked at the United Nations IAEA overseeing their traceability program so formed in 2008 we've been perfectly timed the last two years before being perfectly mistimed the previous 13 so to a certain degree we're ahead of our time but over that time we've amassed significant intellectual activity around databases around the algorithms required to manipulate the databases to be able to use the forensic science to prove where products come from as a company really proud of a couple of statistics so one is our common annual growth over the last six years is 95% so we've been basically doubling every year really proud of that because we only do that from the willingness of clients to engage with us we've got about 160 staff we're looking for another 38 so if anyone's looking for a job we're at the back of the room later but we're rapidly growing to be able to service the demand from our clients a very diverse young company we'll have around about 25 different nationalities of those 160 staff so why do we exist what problems do we solve so obviously forced labour is a central issue today by knowing where your product comes from you can address these issues and we try to to to to to to to where it's forced labour where it's forced labour where it's forced labour and steady and also authenticity so we steady and also authenticity so we help type of thing so we broadly apply our science not just for forced labour but for any issue where provenance is central. So traceability is a word that we all are very familiar with. We are slightly different on that so a normal traceability system will follow a product through a supply chain. Where we're different is we unlock the intrinsic properties of that product and match it back to an origin. So I could take the shirt that I'm wearing now and see if the cotton comes from California where it comes from China, India, Pakistan, wherever. So obviously that's got a massive relevance when we talk about issues of forced labour around the countries Uzbekistan, Turkmenistan, Xinjiang. It's not just Xinjiang. The weakness of a traditional traceability system is it's only as good as the weakest link. So if there is a breakdown for any reason then the whole system the whole system falls apart. So how does it work? I need to caveat I am not a scientist but we will have scientists here today if you'd like to talk to some of our team but inherently we are and we are a product of the environment that we're in. So for example the scientists could take my hair, take a little bit at the start of the hair and see that a year ago I was in New Zealand. More recently I've been in Switzerland so this is this is actually how it has been applied in forensic cases and why is that? It's because every environment is slightly different. So in the schematic here we'll look at the soil. So the strontium in the soil will be different depending upon the area. The hydrogen isotopes will be different depending upon the altitude. The sulphur will be different depending upon the distance from the coast etc. So it's not a black box solution. The beauty of what we do is it's hypothesis based. We understand why fingerprints are different so that makes it incredibly powerful. Not only can we differentiate the fingerprints but we can say why they're different. So trust but verify it's very appropriate being in the Ronald Reagan building who made that quote famous in 1987 when he spoke to Gorbachev said trust but verify that's very much central to what we do. Once we have a fingerprint established we can verify that anywhere along the supply chain which is really really powerful. So again the shirt here we could buy it at a retailer. We could take the sample in the manufacturer. We could take the sample from the cotton. So once we know what the issue is for the client we can put accountability into the supply chain at any point. So Kit was mentioning I think the last question who is responsible. So the important record has the issue that their product will be stopped if it has a problem in the supply chain. This system allows people to put that accountability back into the supply chain. So a little bit more about what we do. So it's a product test. It's a physical test. So we actually test the product itself. The beauty of that is once the databases are created it's very very easy to operate and it's very very compatible with people's manufacturing systems. We're not in there all the time spraying something on or we're not in there auditing every single every single transaction. We can just spot check the product itself. The strength of what we do is driven by the database. So the database has made up a lot of different things. So yes we have the physical samples. We know what their chemical make up is. We can create the algorithms for the fingerprint of the product. But the database is a lot deeper than that. So what we can tell you is the relative risk of a manufacturing site in Bangladesh versus Vietnam is X. Or we can tell you that a T-shirt versus a shirt the risk is Y. So there's a lot of exhaust data on what we do which creates a lot of really deep and rich insights into where the risk lies. So these are powerful for regulators, for companies, for the brands, for the manufacturers. Most importantly it's commercially proven. So we learned very early on that scientists can do anything except make money. So what we have done is we've combined, sorry scientists in the room but I have to be said, so we've combined forensic science, we've combined data science but most importantly we've combined commercial knowledge. So it's no point having a product if the industry won't take it up. So we've put a lot of effort in working with other service providers, working with our clients to be able to get a really commercially valuable solution. So who do we work with? This technology has been adopted in over 25 different products in over 20 different countries. Many of our clients aren't on public record, these ones are. We work with some fantastic industry bodies such as cotton USA who are looking out to protect the industry, the US cotton growing industry that likes the subpoena. So we can work with an industry body, we can work with a regulator, we can work with a brand. And it's not just about the negative, it's also about the positive. So there's a great example there with Primark who put a huge amount of effort into their sustainable sourcing, sustainable cotton program which is all based about empowering women and farming. So the science can be used to detect bad but it can also be used to prove good. So we've got a wide range of clients, but as I say, many people prefer to fly under the radar. So what do the clients say about it? I think the interesting one there is the results are useful either way, right? They either confirm what you thought and what you believed or they don't. And both lead to next steps. So this is really important. A lot of people wonder okay, these supply chains are really difficult, they're really murky, they're very deep and they're very complicated. What do I do if I find a problem? So we work really closely with clients to provide the knowledge to give them the power to be able to make decisions. So it's all about being, it's better to operate from a position of knowledge than ignorance. Once you know you've got a problem, you can start to make changes. You can start to change supply chains. You can start to test the supply chains more. So it's all around empowering people by taking that scientific data and turning it into an insight of which they can take an action on. And look, none of our clients are perfect. We know that. They know that. Everybody is working to improve. And like, it's all about progress to be able to eliminate these problems out of their supply chain. It's not easy. It's really, really difficult, but we can put a person on the moon so we can surely know where our cotton comes from. It can be done if you want to do it. It just takes time. It takes combining the likes of Karen, like Altana. There are a whole lot of different tools that need to be used. And like all tools, it comes down to using the most appropriate tool for the job to be done. So I just want to run through a case study. So people say, like, how do people actually use this in reality? So this is an example of a big box retailer. And what they've done is they've mandated as part of their supplier cotton sourcing standard operating procedure that people must use at testing on the finished product. So they've mandated that to their suppliers. So to bring that to life, we work with a third party agent. And that third party will independently sample, will collect samples through the supply chain, according to a really strict chain of custody and standard operating procedures. So the beauty of that is if there's a problem, we can go back down the supply chain and find out what's happened and why. Then the cotton origin verification testing's done on any item containing cotton in this example. As Kit was saying, like, it's not just cotton. We do tend to think about this specific issue around cotton, which is the largest one, but this could be any product. So this is across private label and also the national branded items. The samples are submitted for testing against two things, a claimed origin and or a prohibited origin. So, for example, they might want to know whether this suit is made from genuine US Supreme of Cotton. So that's testing against a claimed origin. And if we test it for that and it's proven to be from the claimed origin, it's fine, it's clean. If not, they will know is it from a prohibited origin. So then we test against that. The most important thing at this point is if it's consistent with the prohibited origin, the goods can't proceed. So a corrective action plan is then put in place, the products are stopped, corrective action plan is put in place, and then the problem is remediated. We will then go back in the future and test to make sure that that has happened. And we can also help the client by doing pre-production testing at tier two and beyond. So that ensures that the issues are identified early on and that we can find the problem early and remediate. So that's an example of how the industry are using the solution in practice. With that, happy to open up for any questions. Really identify that it is a very complicated issue. It's a really tough challenge that people have got to face with to fix. It's not easy. We're not the entire answer. We are very, very we refer to it as narrow and deep solution. We're very specialist in what we do. It's appropriate sometimes, not all the time, but by combining with other leading technology, we know that we can help solve the problem. So happy to answer any questions. Hi, Chandra Navarro from Hogan Levels. I have a question about your cotton testing procedure. When you're testing a bale of new cotton, how are you testing that bale? Are you finding specific points in the bale so that we understand how accurate that is? And second, how about recycled cotton? And where is your science on that? Great. I'll answer the difficult one first. So recycled is problematic. It's not easy for us. We're looking at technology to do it. We've made some really good progress on blended cotton. So cotton blended with polyester, nylon, etc. We've got that sorted. The next challenge is recycled. So right now we don't have an off the shelf solution. So in terms of where would we test in the bale, so we will work really closely with the client to design a testing programme. It's not even just at the bale, it's like where we test in the supply chain. We've developed an algorithm to allow our clients to create a statistically defensible auditing programme. So based on prevalence, criticality and detectability, we can tell clients how many samples they need to test at what point. By way of example, it's a sample population question. So if somebody was to audit your financial statements, they would take a subset of your statements and that would tell you the health. Likewise if you want to know an election result you'll poll two and a half thousand people to tell you what two million people will do. If you design it right, the sample is very accurate to the population. So our data and stats team developed that for our client. So what we want to do is make sure we don't have any bias to your point about where we test. We make sure that we're doing it in a random scientific way, but really happy to talk about that offline. Hello, I'm Tim Roddy. I work for Vera Bradley on the IT side. Are you working with CVP to create a certified database of all this information so that the border agents can actually use it to clear our shipments? Because that's our problem, is how do we get this data into the system that then it makes the agents job much easier to clear and avoid a WR of? Yeah, great question. So I think what we can say is it's on public record that we have a commercial relationship with CVP. I'm hugely proud of that. In terms of databases, we are developing a portal for all clients to be able to access it. So we'd love to talk to you offline about what that looks like, but that is our problem. We want the technology to be used as widely as possible. Lauren Chevelier, thanks for the great presentation. I had a question about, you were mentioning earlier, sort of spot checking different items in order to be able to use your technology in order to trace. And I'm wondering how that works within the context of CVP's new FAQs, where they talk about DNA and isentropic testing and how you need to be able to make sure that what is being tested actually can be connected to the item that was specifically detained. So do you help importers in that process to make sure that those two connect and sort of what does that look like, thanks. Yeah, great question. So one solution that we're working on there is that a certificate of origin can be attached to the product that has got security features and so it can be proven that the certificate on the bail or the shipment relates to that shipment. It'd be easy for someone to create a counterfeit or attain certificate and then we're worth nothing. So yes, we do have a technology. It's not rolled out commercially yet, but we've developed the prototype. So yeah. My name is Jeff Wheeler. I'm the Director of the Global Trace Protocol Project. My question is you said that you're able to check whether the cotton comes from a prohibited location or source. So the question would be can you identify the cotton that's coming from Xinjiang, China? Yeah, that's the service that we provide. Hi, Caroline Dale from Flexport. Could you speak to any other commodity types that the testing is currently in commercialized and prepared for other than cotton? Yeah, great. So sure. Cotton will lead the red meat coffee. We're working on timber, we're working on palm. We're working on some extractives such as mica, polysilicon. So yeah, it's broadly applicable. The underlying science is largely the same. There are nuances. Yep. So there are a few. It also comes down to a resolution. So we can deliver a solution to like a farm of origin, a region of origin, or a country of origin. So we've got different resolutions. We also have different ways of asking the question. So we could say does it come from California? Does it not come from India or where does it come from? There are three different questions and there are three different algorithms. So and all of the time the databases are different. So yeah, there's a bit of nuance to it, but we're building them. Yep. Hi, Christian Roseland, Clean Energy Associates. You mentioned polysilicon there. How do you look for the impurities in a very, very highly refined product such as polysilicon? Yeah, look, that's going through the R&D pipeline at the moment. It is really complicated. We've had some really good progress recently. In the room, my colleague Rupert Hodges, who's the Deputy CEO, has been largely involved in that. Rupert, put your hand up. He can answer some detail about that. That's one of his projects. Thanks. Okay. Thank you everybody. Alright, moving on to our next speaker. I'd like to introduce Ms. Amy Morgan. She is the head of Trade Compliance at Altana Technologies. And prior to Altana, Amy was Vice President for Cross-Border at Avalara, responsible for developing technology solutions to automate landed cost calculation and cross-border tax compliance. Previously, Amy managed global trade and customs compliance operations for Amazon, Costco, Wholesale, Nordstrom, and Microsoft. Amy is a licensed U.S. Customs Broker based in Seattle, where she was recognized as one of the Puget Sound Business Journal's innovators of the year in 2019 for using her big trade background to pursue progressive solutions to compliance challenges for companies of all sizes. Good morning. I'm Amy Morgan. I head trade compliance strategies for Altana. Did I do that? Altana is using artificial intelligence to build a dynamic shared intelligent map of the global supply chain so that governments and companies can connect their data to see their multi-tier extended supply chains and to illuminate their multi-tier supply chains and to facilitate trusted trade activity, including the elimination of forced labor. Before joining Altana, you heard my bio. I was a trade compliance person. I managed trade compliance for some large U.S. importers, just like many of you in this room today. And the UFLPA did not exist back then when I was managing trade compliance, but I still remember how painful it was to obtain that outer-tier supplier data in order to respond to CF28s or to do origin verifications or to file OGA declarations. I remember sending countless supplier emails. I remember sending reminders and questionnaires and surveys. I remember getting responses and then sometimes not getting responses. But all to trace the inputs of the finished goods that I was actually importing. So this is how I know that compliance with the UFLPA is going to require artificial intelligence support. Our traditional processes as trade compliance people just simply can't scale and besides, compliance with the UFLPA is going to require more than simply mapping your extended supply chain. It requires more than simply mapping risk data. Static, manual, questionnaires and surveys and audits alone are not going to get us there. So Altana is using AI to build the only dynamic map of the global supply chain, a source of truth that we call the Altana Atlas. The Altana Atlas is made up so we pull together sorry my notes are sliding right off the podium. The Altana Atlas pulls together data sources from public sources, commercially available sources and non-public sources, first party data. We fuse that together, we clean it, we standardize it so that that makes up the map that I mentioned earlier. But it also learns. So at Altana we use federated machine learning models to pull all that data together so that the map itself learns from all of that data, public, commercially available and non-public, so that we have shared intelligence without actually sharing sensitive user data that's really important. The result is this dynamic map that reveals multi-tiered networks automatically and enables and encourages collaboration across all supply chain stakeholders. The Altana Atlas is alive. It's dynamic, meaning it evolves as the world around us evolves. Our map is always on, it's constantly updating as we add new transaction records weekly, we add new company information as companies move, close up shop, change names, change addresses, or as new entities are added to the entity list or removed from the entity list, and as new WROs are issued or revoked, the Altana Atlas likewise responds. Now as a trade compliance person, and some of you are supply chain managers as well, we are now required to know about the non-compliance that might be lurking way upstream of our value chains or downstream in our supply chains. Now this, I don't know about all of you, but this is visibility that I don't believe we are equipped to handle. We never had to handle it at scale, right? This is exactly the sort of visibility that Altana offers. The Altana Atlas has done the hard artificial intelligence work, so all you have to do is use the data you already have to get the insights that you need to be compliant. When we onboard our customers to the Altana Atlas, we ask that they provide their supplier data, all right? Once we have that supplier data and we link it up to the Atlas, this situates a company or governments own supply chain in the global network, the grand scheme of things. Now we can see beyond that tier one, we can see beyond what you already know, without having to issue questionnaires or surveys or any crazy emails or not getting any responses, you get to start from a place of trust. And it's only in my opinion, and I believe, that it's only with this sort of network connectivity that we can derive the risk insights that we actually need to be compliant with the UFLPA. This same information is likely to be needed for other net new policies and initiatives we've heard about that are lingering on the horizon, calculating greenhouse gas emissions. What about forced labor regulations beyond the United States? Deforestation, conflict minerals, increased sanctions, same information is needed for all of those. I can even imagine a world where this same sort of network connectivity is required to help with the increased enforcement of current trade compliance priorities. Think about validating the countries of origin, identifying possible transshipment or being able to predict what is and isn't a counterfeit product. So just imagine what if you could see your entire supply chain network without painful effort? What if you could use artificial intelligence rather than manual screening to surface exposure? Imagine shrinking the haystack so that you could focus your expert resources and audit resources and investigation resources on only the most at risk supplier activities. And finally, imagine that you are able to engage with your stakeholders at a meaningful level with real data rather than relying on them to come to you with that information. So if this next slide works, I'll be able to show you an actual video. I did record a short demo, quite high level, but let's give it a whirl. This is the Altana Atlas in action specific to forced labor. In this example, I'm a supply chain manager at Uniclo, an international apparel company that sources textile items from factories in several countries and imports them all over the world. Today I want to know whether or not the manufacturer of any Uniclo products has potential ties to forced labor activity. This effort would normally involve a long and expensive project, but with the Altana Atlas, I can initiate this investigation with a quick search of my company name. When I search the Altana Atlas for Uniclo, I'm presented with a list of search results much like a Google search. And I can see the top result for Uniclo company limited is likely the most relevant. Clicking into this result takes me directly to a knowledge graph view where I can explore Uniclo supply chain relationships. In this view, Uniclo is the top node on the graph, while the nodes below represent companies that have either sent to or received shipments from Uniclo. Those shipments are represented by these links or edges that connect to the nodes. Uniclo is a large company with a lot of global trade relationships which can be overwhelming. To make the review more manageable, I'll add a filter to reflect only those relationships with the most recent activity. This reduces the number of entities in my network making it easier to continue my analysis. With this narrowed view, I can explore any of these supply chain relationships to learn more about who these companies are, who they trade with and if any of their relationships could pose a potential threat to my business. For example, perhaps I'm interested in the supplier Geoco Danoa. When I click on this company, a summary of their business appears on the right sidebar. And when I click on the edge that links them to Uniclo, I see the number of shipments transacted with access to a table of detailed information about those transactions. I can also explore which other companies this supplier is engaged with and look for any potential risk flags in their network that could be harmful to my business. However, to determine the sustainability profile of a company, it's not enough to simply understand their business relationships. So not only has Altana constructed Uniclo's extended supply chain relationships, but it has likewise constructed the more granular multi-tier value chains that make up those relationships. With one click into the value chain screen, I'm presented with Uniclo's multi-tier value chains. In this view, Uniclo's company facilities are represented as the tier zero. The tier one represents Uniclo's direct suppliers. These are largely the manufacturers we source finished goods from. In tier one, we can derive their suppliers at the tier two and their suppliers at the tier three and so on. For the first time, I have visibility beyond my direct tier one relationships. And this view traces the flow and transformation of inputs from raw materials to the finished product linking manufacturers to their suppliers and their suppliers so I can see beyond the entities producing Uniclo garments to those suppliers who are making the fabrics and spinning the yarns all the way to growing the cotton or extruding the synthetic materials. And it didn't require that I manage any manual questionnaires or surveys. The Altana Atlas did all the hard work for me. Most sustainability goals like climate, carbon, natural resource consumption, regulatory compliance or human rights will require starting from this granular view in order to see how their supply chains align with relevant commitments. But because I'm focused on forced labor today I'm going to filter these value chains for the specific risk type that we are looking for. This filter works because not only did the Altana Atlas construct Uniclo's multi-tier value chains, it simultaneously screened the entities in those chains for potential forced labor exposure. Right away I can see there are several Uniclo chains that may be compromised as indicated by the warning flags visible here at the tier 2 and 3 level. To make this view even more manageable I can filter for a specific supplier let's say Shensen so I can focus on understanding how this company out at the tier 3 level may be relevant to my business. Altana has made it easy to connect these risk dots. By clicking on the warning indicators we have access to an explainer that tells us Shensen wa fu was flagged because they are linked to companies located in Xinjiang China which is a region subject to forced labor restrictions. This explainer also provides links to the long-sourced citation which can save hours of manual research later. I hadn't seen that one in a while. It's weird to hear your own voice played really really loud. Back to you. I hope that demo inspired some ideas around how you could also use the Altana Atlas to comply with the UFLPA but over the break somebody asked me if I would provide a few of my own favorite use cases just inspire additional thoughts. So I said sure so I jotted a few down. So first of all piggybacking off of what the previous two speakers said this sort of insight would allow you to take early action to provide possible to identify possible exposure sooner. So the sooner you can start investigating right and gather all that evidence the better because once there's a detention it's often too late. You can also save these insights as part of your record keeping program. This way you could prove how exposure may have been identified but was later addressed. On their own these sorts of insights are not going to satisfy the admissibility package but they could help you construct one. So if you had access to this sort of a value chain now you automatically know it's like a blueprint. I know who I need to go collect which documents from. Making it a little less what I say a little less shot in the dark and more paint by number. Now you might also consider proactively communicating any identified exposure with CBP. Or maybe in support of your CT PAT obligations maybe as evidence that you're actually going above and beyond to exercise reasonable care with the forced labor minimum security criteria. You can also maximize your compliance investments by sharing these insights with third parties like in or attain to optimize the investments you're making. So perhaps you're not auditing everything you're auditing only those things that have a likelihood of being exposed. And finally it's also possible to layer in additional data. You heard from Kit and Grant earlier. They all Keron offers data sets Moody's offers data sets that can all be layered in because it has created this map adding additional data elements. I say it's easy but Peter over here my colleague is probably going to tell me it's not that easy but it's so easy it's just data right you just plug it in and it works. So if the regulations continue to assert that exercising reasonable care means taking an active role in knowing who your extended supply chain partners are incorporating is the only way we're going to do that at scale but fortunately we have the right technology at the right time in the world and with that I'm happy to take questions if they are technical I will make Peter answer them. I'm a trade nerd not an engineer. Good to see you. Your interface looks awesome it's so impressive I have a couple questions so I'll probably need to talk with you afterwards as well which is a benefit for me. I guess one question I have immediately is just which data you're using to actually conduct the screenings you mentioned the kind of ability to screen suppliers there I'm wondering is that like an exager or a Keron. Okay great question so to do the screening part we do incorporate all of the international restricted party lists that money can buy we have incorporated we've also incorporated the entity list all the WROs the Sheffield report and then so that's what makes up the screening currently as far as the Keron or Exager or any of these other data elements those could be layered on as additional we welcome that actually we all work very well together that is the intent I think I answered it was all just about the screening right right so if a company is already using Exager to conduct their screenings they could potentially work with you guys to layer that on as well is it that easy? Peter says yeah it's that easy it's just data right? I guess the second question I had and then I'll cede you mentioned like other companies inputs plugging in so obviously a company comes in they have the blueprint and then they can kind of maybe do some of their own diligence to confirm the supply chain is accurate for one of their products right so would their data then be in your system and available for other companies to view? so no is the short answer what you saw was a very high level demo I only had a very short period of time right we could show you a more in-depth demo where at risk or likely exposed value chains or supply chains could be you have the control to verify them you could say in review verified they've been cleaned or this is high risk and I want to make some better decisions about what I do we could show you that so the controls are now whether or not other people can see it no the way we deploy and again Peter step in please because is every client when we work with them all of their data lives in its own we call it a spoke so it's a virtual private environment where their data goes their data doesn't leave and their data nothing goes in nothing comes out actually one of our big uniqueness is that we deploy our data to the technology we don't ask you to put your technology into our data so we call this when I say federated learning it's we take the intelligence from your data not the data itself but the intelligence from that data comes back to feed the graph so the graph is constantly growing and evolving and improving which makes your results a lot more accurate but we're not sharing anybody's sensitive data did I get that okay no that was exactly right I mean we built out this private enclave system to enable sort of this risk management while respecting data privacy sovereignty and security so have your cake and eat it too yep awesome thank you very cool Hi Caroline from Flexport Virginia hit my first two questions so that's perfect so from the knowledge graph view I'm assuming that that's reliant on both sort of customer data and public shipping data to get to those shipping partners how do you deal with issues of manifest confidentiality or modes of transportation there this question has come up five times already this weekend it's only Tuesday so we that data that we purchase that that commercially available data that we acquire that anybody could buy right I don't know actually maybe Peter this is a question from yesterday I can just ask you I don't know if when you have vessel manifest confidentiality if that data is redacted or just left out of that data that we purchase I don't even know if you know the answer I can actually speak to this that's my pleasure so yeah thank you for the question so first of all start off by saying we're very proud of our knowledge graph it's the largest that we know of it's not perfect there are things we can see and there are things we can't see what we set up is so manifest confidentiality for example we have two approaches dealing with that one triangulating through multiple data sources so maybe redacted or otherwise censored in one but available in another two the federated knowledge of the knowledge the federated nature of the knowledge graph by which I mean through its multiple deployments it can access proprietary and first party data that gives more signal and then three we have probabilistic methods which are not displayed here I mean we've shown you everything that's deterministic but other approaches to sort of rate risk and deal with redaction that we continue to develop so I'm happy to speak to that latter point in more detail at a later time but really the desire and what we've gone through here is to build something that brings you the best information available from both first party proprietary and commercial data that is dynamic updating to multi-tier and then allows a flow to go through and investigate any vagueness or uncertainty in a rigorous way and always gesture a little bit towards that interface there so thank you for the question I hope I answered thank you Hi Amy, Jasmine Martel from Hush Blackwell I have a quick question on the network connectivity that you're building one of my questions is in regards to an industry perspective what does that look like as far as your users and are you seeing that a lot of your data is more so gained from like the apparel industry or users in a certain area or is it even as far as the information that companies are receiving so I can take the users if you want to take the coverage so as far as users go we target three user types so we do target government we remiss if I didn't say CVP or any other customs authority is certainly a target market for us we also target logistics service providers so we have a very public, very deep relationship with Maersk for instance where if you are a Maersk client and you are interested in what I showed you today reach out to your Maersk folks because we have a relationship with them specific to forced labour but other logistics service providers who want to offer similar value added services for forced labour visibility is definitely right up in our space and the third market would be enterprise a lot of the companies in this room are potential clients of ours are already clients of ours so that's where we spread out across those three users but as far as data coverage Peter I'll let you take it yeah so thank you for the question so speaking of data coverage we are actually covering all physical goods and in general all modes of transit there are different areas where we have even more data and areas where we have less and this is a function both of our data sources, our deployments, all of these things but I will note that in principle it's covering some of these data sets are the scale of entire economies in fact many of them are and as such it's widely applicable cool well you can't miss me so if you have other questions please find me and we do have a table in the back but thank you for inviting us all right everybody three quick reminders before we break for lunch first the CBP team is eager to engage with you so if you have any questions please find someone with a green lanyard and ask away second please stop by the rotunda during the lunch break to meet our team members there we've got the CTPAT team we've got our human resources team and we have several companies that are eager to talk to you there and finally we encourage you to use the continental rooms which are shown on this map here the red dots to you know meet with people and have discussions there with that we're going to take a lunch break and be back here to start at 1pm thank you very much all right everybody I think we're going to get started now just a reminder that each of the companies presenting will speak for about 15 minutes and then we'll have 10 minutes of Q&A we've got the microphones right there so you can queue on up when it's time to ask questions but we also have somebody who is roving with the microphone so you can always call them over and speak that way if you need to so let's get started with the first introduction we've got Dr. Brett Tipple so Dr. Tipple is a chief scientist of Flora Trace Brett is a substantive expert on geographic origin tracing and authentication methods during his career he has supported law enforcement in solving cold cases helped identify and repatriate US service members remains and combated food and beverage fraud in the private sector Dr. Tipple thank you and thanks for having us it's a great honor to be here and we appreciate the invite yeah I'm Brett Tipple I'm the chief scientist and president of Flora Trace we seek to help companies big and small manage the risk of hidden force labor in their supply chains at our core we're a scientific company that applies leading-edge technologies to determine the origin of organic and geologic materials we utilize forensic chemistry and data science approaches to assess where a material was grown or produced these technologies support by helping them demonstrate their materials did not originate in Shenzhen wholly or in part and they play very well with a lot of the other solutions that we've heard about this morning so if you're detained under the UFLPA an importer may claim that the UFLPA doesn't apply to them as their goods did not originate in Shenzhen for the applicability review the importer will be challenged to provide evidence of geographic origin the question is how do you scientifically and quantitatively demonstrate that your goods did not originate from Shenzhen so consider these two piles of tomatoes here can you tell which pile came from Shenzhen and which one came from say India I know what you guys are all thinking those two piles are exactly the same I can't see a difference and that is enforcement's perspective as well one tomato looks just like the other tomato so is there a way to distinguish this pile of tomatoes from Shenzhen from this pile of tomatoes from India when they look, feel, taste smell exactly the same what about the individual tomatoes within that pile was there any commingling between the different growing regions in that pile of tomatoes what about the tomato paste that finished product where there's hundreds of individual tomatoes that go into that how do you know that that starting material was Shenzhen free by looking at these tomatoes these questions seem impossible to answer but in fact you can answer all these questions because organisms naturally record their geographic origin within their tissues the innate chemical characteristics of all living things actually reflect how and where it was grown here's an example of staying on the tomato theme a little baby tomato plant this tomato plant will build its tissues the carbon, the hydrogen, the oxygen that's going to incorporate in these tissues from the water that's being supplied to it from the atmosphere that's taking in it will incorporate elements and other isotopes from the soil and the fertilizer that is being fed since these chemical characteristics depend on where and how the plant was grown these variations provide a unique chemical profile or what we call an origin fingerprint of geographic origin so consider that most of us here in the room can grow tomatoes in our home garden wherever we live the innate chemistry of each of those tomato plants are individual tomato plants reflect where we grew them and how we grew them did we take care of them did we leave them for that week when we went to vacation that's going to be reflected in the chemical characteristics of those plants so each of your little tomato plants will have a unique origin fingerprint associated with it our company floor trace measures these origin fingerprints and this allows us to determine the geographic origin of materials and this is how we help importers demonstrate the origins of their products so how does this work in a commercial commercial good in products so we work with companies their suppliers to first identify the unique characteristics chemical characteristics and define an origin fingerprint for their product since the origin fingerprint is naturally embedded within the material itself it can be used throughout the supply chain it can be sampled at any point to confirm the origin of that product with a well designed origin testing program it can be demonstrated that a product was not minor produced in Shenzhen using this technology the importer can include this sort of information to support their claims of a UFLPA applicability review so in practice we help importers better understand their supply chains and enhance their sustainability sustainable sourcing practices the steps in this process first include an understanding of the type of product and how it's grown and where it's grown and the type of environment it's grown in next we'll go about collecting authentic materials we'll analyze them to determine a baseline a chemical baseline the type of data we'll collect there's isotopic values elemental abundances the relationships between these as well as a lot of other metadata associated with the growing region and other parameters during this period of time we'll also identify and collect other authentic materials from outside this very specific region we're going to use those to compare our our authentic from the region we want to define then we're going to dig down into that data and we're going to find that unique profile the chemical profile and from there we're going to identify and isolate a origin fingerprint for that targeted growing location we use these origin fingerprints to verify the origin of the latter products and we compare these data with the other data that we've generated from outside that specific region so that way we can say it is versus not from the very specific region that we're claiming it's from we use a wide variety of statistical methods to quantitatively demonstrate that the targeted good originated from the specific growing area and not Shenzhen here's an example of the type of clients we might work with here is a medium size a consumer good company importing onion products from India into the United States they have a really great relationship with their suppliers they regularly visit the facilities and visit the production sites so their suppliers indicate that they only purchase Indian onions while the importer trust their supplier they know that about 25% of the global onion supply is grown in China with the majority of that grown in Shenzhen this is a proactive importer they understand that onions like this are going to be at risk of forced labor and likely to be targeted they understand that a lot of the Shenzhen onions are being shipped into India to be processed we worked with them to demonstrate that the imported onion products were not produced in Shenzhen they not only used the outputs of the origin testing program as evidence in the event of detainment and applicability review but they can also demonstrate to their clients that their products do not originate from Shenzhen and are free of forced labor so our science has been pioneered in collaboration with law enforcement and intelligence community my background is in forensic science this came out of a lot of high level research and development projects with the federal government our background is in using this technology in a lot of legal contexts we're very rigorous and our data is reproducible some of the things that we've worked on you know sourcing anthrax spores back in 2001 here in DC counterfeit currency as well as understand the origins of narcotics and other illicit drugs identifying and linking where explosives are made it's important to note that our solution verifies the geographic origin with natural chemistry we're not spraying anything on it it's labeled naturally by the environment we're testing the actual product itself and it cannot be replicated these origin fingerprints cannot be replicated compare that to some other solutions out there genetic DNA solutions they're going to identify the strain or the species but you can grow a particular species at many different locations and then against paper or label those are only going to track the packages not the actual materials inside our technology actually tracks the materials itself so some of the things about what we offer I talked a bit about our origin assignment verification program we also do growth condition verification that is organic versus conventional fertilizers as well as products that have been grown indoors or outdoors areas that we work in around these concepts are around ESG as well as protecting specific intellectual property for a product that might have a unique characteristic based on where it's produced also do analytical testing we're very good at research development if there's a problem we can help design research around that to get an answer to that we also have advisory services to help clients understand their risk and answer some of their questions about our tools so our expertise and our pipeline here focusing on the UFLPA we work a lot with foods, beverages agricultural products as well as geologic materials as well so why floor trace we're here in we're a domestic laboratory our laboratories are in Utah given our background in the legal and government work we have very good chain of custody and secure storage protocols as well as that our data stands up as evidence in the court of law meets the Dalbert standard it's been challenged multiple times and has met these challenges we strive to do fast turnaround and be as cost effective as absolutely possible so I can stop there and I'd love to take your questions and we will be back at the table in the back so please come up to the microphones Hi, Cindy Delion with Delion trade thank you for your presentation can you go into a little bit more detail into the chain of custody piece especially in the context of a CBP detention notice yes so we we have protocols in place for chain of custody you know we have all the the the signature forms and things like that so we would provide those sorts of documentations envelopes and things like that that would have be sealed provide those in the event of a detention that's something that you know we're we'll be working on for that actual application so that would be something perfect thank you alright so we're ready for our next speaker Mr. Michael Pro Rock he is the founder and CTO of measure.io he is a highly accomplished technology executive with deep expertise in machine learning analytics and decentralized systems I'm told he also has a passion for outdoors photography and woodworking and resides in a small town on the border of the greater Yellowstone ecosystem which I'm personally very interested to hear more about later you're up thanks so much yeah so as mentioned I'm out in that Montana direction and I got to leave beautiful 50 degrees sunny weather to come out here and possibly get some snow so we'll see how that all goes I wanted to start by just talking a little bit about some quick company background and how we got from kind of starting the company back in 2016 living out small farm country in North Carolina to today and really I started this company with a mind of working specifically at the farm level and to look at data across the board from I can't tell it's actually behaving now you know started working primarily at the farm level to make predictions and recommendations out to farmers that led us to building up a whole bunch of not just environmental data but to then perform automated machine learning on top of that environmental data to understand what was actually going on at the field level how much growth could be occurring what diseases were present how much could be present in the pipeline at any given time we expanded from there when COVID hit into applying our technology to start broader tracking disease kind of spread outbreak impact all sorts of fun stuff there not just on the crop side but also on the human side and then from there started broadening our systems out because they work on unstructured data broadly to go ahead and look at trade information so today talking about forced labor a couple of key problems there with forced labor we've got really specific problems mostly in the fact that the data is all over the place even the data that we see that's been highly qualified and heavily reviewed so we're sure it's a problems in it and it changes all the time the other aspect to ours because of that we wind up with huge margins for error in terms of data collection and assessment when we look at things like physical testing and tracing it's important it's a key part of it but it's expensive and even there when we look at items like cotton where you have inherent processes like lay down during the milling process where cotton from different regions is blended to actually get a product you can work with makes it really hard to trace down each individual fiber additionally basically no products that we deal with at the consumer level are a single product even something like this shirt probably has five or six different manufacturing steps going in not just cotton from different areas but the collar might be made at one particular factory and then assembled at a cotton sew facility in India or elsewhere so why did we go ahead and build the earth stream platform we went ahead and put it together largely as mentioned first start dealing with climate risks and what the impact was going to be at the farm the other side was to actually go through and start a process of building out an actual digital twin for every component that we're looking at worldwide so every single farm every single actor in the supply chain every single entity as they change names or relocate and firmly to base all of this stuff on hard science really starting back at the biological and environmental modeling side based on field data and collecting that information continually analyzing it but then also recognizing the fact that we didn't want to get into a chat GPT situation and have to keep placing band-aids on the outputs of our models we wanted to build the ability for scientists to actually put logistic inputs into our system at the front end side and review data before it actually goes out for recommendation purposes so how do we go ahead and attack the forced labor problem we tack that a couple of different ways but we really do start at the source we're continually going through and monitoring information what crops can be grown where what can be mined where and actually going through and looking at very similar to what we've heard from some other folks right because there's only kind of one common sense way to do this to go through and actually follow that stuff outbound from the source when we then intersect that stuff with customer information we encounter on the web we trace backwards towards that source and let those two graphs intersect and we do all of this with the process of continuous learning and continuous information gathering we don't actually send human analysts out to go out and collect a bunch of information we don't have room to soil people we've built a bunch of models to do this stuff for us so how do we do this from a data standpoint we start literally at the imaging and remote sensing standpoint we're looking at everything from synthetic aperture radar to actual growth conditions in the field and mapping that on an ongoing basis and continually improving models for processing that data and intersecting it with the unstructured data that we find on the web that say might be an assessment report from Sheffield or otherwise we go through and actually break apart and understand when we identify cases of forced labor what economically is driving that and creating the situations on the ground that are actually leading to the benefits of forced labor there obviously there are genocide aspects as we do see in the case of China but there are broader things there the reason it comes into play from a forced labor standpoint frankly is because it's economically advantageous to the PRC we also then are continually tracing every piece of information we can find on the web and continually expanding our crawlers for information that take unstructured data break it apart and automatically link it together and qualify it with other information we've already encountered what does this look like simple dashboard standpoint we customize this look at feel-wise, logo-wise, etc for our customers and we turn on modules as may be appropriate we're looking at a trade dashboard here looking specifically at some cotton output and some anonymized media that's coming in but we obviously have different dashboards in play when we're looking at things like disease outbreaks or other you know risks of say closures and things like that we also though have this notion that we're never going to know everything and that we have to have a way for users of our system to go ahead and direct the system as to how it should utilize the machine learning and algorithms built into it so users can upload their own documents in any format we take those documents as they are identify every company in them every entity, every product every organization and then use that to further information gathering we cross reference everything geospatially we're not just looking at interior mapping, we're going through and actually trying to physically map all of these items in addition to who is supplying who and then continually identifying additional information as it arrives in real time in order to present before an analyst so that they can make appropriate decisions down to the farms, the mines the oil wells where the raw materials are extracted the good news is we've had some time to work on this and in the past 12 years at SourceMap we've been able to help, right now there are more than 250 brands that use SourceMap to map their supply chains and we're able to get all of these industries here to map down to the raw material origins so the good news is supply chain mapping is here, it's eminently doable and in fact hundreds of companies already do it but it's not evenly distributed and so I'm going to talk to you about that now supply chain mapping is also why do people do it it was a sustainability priority for many years it only recently became a trade compliance focus so it is in fact used SourceMap to manage risk of many kinds, resilience sustainability but also compliance with programs like CTPAT which is here today with various European due diligence laws and then finally with forced labor due diligence section 307 and the UFLPA in fact time magazine named SourceMap's forced labor due diligence solution, invention of the year 2022 when it was launched in June last year okay what makes SourceMap unique so section 307 forced labor enforcement in general is extremely broad it's global in application and it covers all industries and all raw materials universal supply chain mapping software every company buys a combination of mind and farmed and synthesized raw materials and turns them into proprietary products that follow unique bills of materials or recipes so what you need to map a global supply chain and ensure there's no forced labor is a universal supply chain mapping tool that can be sent to all suppliers in all countries all at once to get all the way back to where the raw materials come from we verify that data the data submitted by the supply chain through continuous tracking of the transactions and that means the receipts the bills of lading we continuously verify the supply chain data and that means cross referencing that means looking for fraud that means looking for counterfeiting adulteration every step of the way and we do all of that in highly secure databases that are managed by our customers because this is all very sensitive data and this is not data coming from the public domain this is actually every company's own supply chain its own formulations its own transactions down to raw materials and then last but not least we've transformed a handful of industries into being transparent but there are a few that are left and so we also have a large team of experts into this new normal of supply chain transparency all in all SourceMap provides basically the full suite of software that a company needs to match the CBP importer guidelines so that means everything from supply chain mapping identifying all of the stakeholders in the supply chain through transaction traceability to verify the supply chain, risk assessment to independently verify the veracity of that data collecting, supporting evidence things like mitigation plans and corrective action plans and then finally reporting just a few interesting tidbits on average a customer of SourceMap discovers for the first time 20,000 suppliers that they did not know that they had through the process of supply chain mapping which I'll go over that's a lot of new companies that our customers discover that they're in business with to take reasonable care and ensure that they're compliant there are tens of thousands if not more individual product codes that have to be uploaded into SourceMap in order to trace all of the bills of materials for the typical customer and more than 100 gigabytes of supporting documents are collected annually so it's tough to do this with emails and spreadsheets and PowerPoint you really need a dedicated database and data collection tool here's what it looks like when all the hard work is done supply chain mapping is as everybody knows it's putting GPS points on every single supplier in the supply chain why because once it's mapped you can audit it so this is the gold standard for supply chain transparency on the right you see a network diagram that's actually the flows of material between actors in the supply chain extremely important to know which items are in an individual container but also to understand which suppliers are still active and then risk assessment which is an overlay of third party data to independently verify the risk of the supply chain supply chain mapping works through a cascading workflow that is extremely effective it basically works like LinkedIn if you're familiar with social networks companies invite their tier one suppliers who invite their tier one suppliers and over a sequence of several weeks that gets down to the tier five six seven suppliers of raw materials so this process starts with SAP or another kind of enterprise supplier database and then it quickly balloons into those tens of thousands of newly discovered suppliers when it comes time to verify the supply chain that's all done through data that can be uploaded asynchronously which is to say you're farming goods on their own schedule before a purchase order has been placed if you're manufacturing you're waiting for your customer to place a purchase order all of these entities are uploading transaction records into source map if they don't have their own inventory management system they can just use source map as an inventory management system and we stitch together the data once the purchase order has been placed and the shipping documents are ready for import verification once you've mapped your supply chain that's when you can start to verify it there are many kinds of verification that we conduct ourselves looking at the data but also in partnership with others who provide very powerful data sets to look for risk we normalize the supplier IDs we use global business identifiers we deduplicate the suppliers geographically we look for risks that are geographic in nature we look for risks in the volumes that are being traded that would mean places where there could have been adulteration places where there could have been counterfeiting and also places where capacities don't match what is being produced so too many things being produced somewhere not enough people not enough machines not enough land we have a mismatch and that's a red flag for unauthorized subcontracting we also partner with database providers such as Keron who you heard earlier today has extremely high quality data about the risk that a supplier in a customer supply chain is affiliated with an entity on a named list or with an entity that is otherwise high risk and we bring all of that information to our customers proactively so that they can put in place those corrective action plans and ensure that none of these risks are actually affecting their supply chain something very important about supply chain mapping when you are doing it with enterprise data is that we've created a standard for vast amounts of data to be shared within a supply chain and so that means that data has to be very carefully shared so there are no data can ever be shared between competitors it can ever be shared between suppliers in the supply chain all of the data is for the benefit of the importer so all of the vendor IDs all of the transaction records those are all put onto local servers that are managed by the importer the data is anonymized and aggregated whenever possible anything we can do to shield personal data in the case of privacy regulations and to shield trade secrets in the case of proprietary formulations for example and we do all of that in the aim of being able to issue a verified by source map check mark which is to say enough data has been collected about that supply chain that we can be sure that it is authentic okay last but not least on how it works supplier outreach so there's a lot of promise in artificial intelligence there was a lot of promise in blockchain none of this works without some significant culture shift inside industry and we've seen it happen for apparel, we've seen it happen for food we've seen it happen for luxury we're seeing it happen for pharmaceuticals in the automotive sector and there are other industries that are just starting without very concerted supplier outreach so we have a playbook, we have a checklist we know exactly what steps suppliers must go through to be introduced to the idea of supply chain transparency and to take part and to see the benefits for themselves and what that does is it achieves pretty tremendous response rates for upstream suppliers reporting is I already showed you but maybe the part that is the biggest time saver other than if you've ever tried to map a supply chain in PowerPoint you know that it's nice when there's a piece of software that automatically visualizes the entire thing for you is we also have PDFs that you can print so if you prefer paper as I know many people do you can actually just print that report for that container and hand it over to whoever is asking we've had the good fortune of talking to nearly a thousand companies in the last three years and we're able to provide each industry with the adoption workflows with the transition plans that are suited to them so if you're coming to SourceMap and you're an apparel company more than 50% of your suppliers are already in SourceMap providing data to your competitors it's a very very quick launch if you're in the energy and renewable space or semiconductors it's a bit of a longer journey and it starts with sharing and planning a communication to your entire supply base as to why this is such an important activity to be part of so depending on where you are Food and Ag you'll see most of the largest brands in the US already using SourceMap today there's other reasons to map supply chains so if you ever say well this is too much of a burden just for compliance it isn't it's too much of a burden I would argue to find out until you run out of raw materials that your Tier 2 or Tier 3 supplier is a single choke point upon which your entire global business relies and so more often than not when companies map their supply chains for the first time they will discover that in Tier 2 or Tier 3 they are heavily concentrated in parts of the world that they don't want to be dependent on or with individual actors that don't have the capacity to meet their growth projections and that's exactly the big value so if you want to be compliant with UFLPA you need to map your supply chain but if you want to sleep at night and know that your company can continue to grow you also want to map your supply chain in the case of the UFLPA the Force Labor due diligence solution incredibly rapid adoption since it was launched in June of 2022 more than 5,000 suppliers already participating in the data sharing efforts towards the UFLPA compliance I think even more than usual every supplier in Tier 1 that gets invited to Sourcemap for FLDD for the Force Labor due diligence solution yields 20 new suppliers that were not in our customers databases before so if you start with 1,000 suppliers you now have 20,000 so incredibly rapid adoption and amazing impact I'll give you one little example in the two minutes I have left what happens well you know you've heard about a lot of solutions that can mine public data for the probable affiliations of your suppliers with bad actors Sourcemap is a system of record so we actually hold the data on the actual transactions that happen in the supply chains our customers are often confronted with probabilities that their supply chains contain bad actors and they come back after looking in the Sourcemap database with a definitive proof for or against those assertions I'll give you an example we had a customer that had a flagged supplier and we were able to show that while it's true that that Tier 3 supplier had done business with a named entity in the past the materials that were manufactured by that alleged contact were not used in any of our customers' products and that's something you can only do when you understand the actual bill of materials the formulation of the products the actual records of transactions and then the supply chain map so it's a very important thing that we use both mind data but also proprietary customer data company data to make sure that we know exactly what is coming from where we have a few projects ongoing with the public sector we are involved in the global identifier pilot through our customers we are working with the Department of Labor on data structures for traceability to prevent child labor together with Verite on the streams project and I'd like to propose to you today that source map since it's such a powerful tool for collecting very large and complex supply chain maps and all of the supporting information that it be considered as a case management solution to really cut down on a lot of the work that CBP is doing and other government agencies to map supply chains using PowerPoint or Excel and then last but not least supply chain transparency isn't just good for compliance and to keep force labor out of the US it's not just good for resilience and making sure supply chains can continue to operate it's also good for all of us and you'll see brands more and more actually making their supply chains transparent to the general public to build trust with consumers so this is not just avoiding risk it's actually creating value and building trust and that we happy to take your questions and also show you a demo at our table in the back oh and scan the barcode hi auditability of what gets into source map how are you auditing and confirming that the information is accurate and are you sending auditors out or are you relying on technology to do that absolutely so the auditability of the data is the reason for the supply chain mapping so we're creating an audit trail for every US bound container through this process the verification happens essentially in two ways there's inward looking we look for the coherence of the data and we look to make sure that every transaction is accounted for so that there are no volumes lost or gained along the chain of custody from raw material to import and then it happens through third party referencing and that's looking up databases like the Keron databases and we look to make sure that there are no any of the supply chain maps entities lists sanctions lists and the like to make sure that neither the suppliers that have been mapped nor any of their affiliated suppliers are appearing on any high risk maps great oh hello Maria from STS you mentioned supply outreach so you can fill out the information that you're looking for supply the documentation when this is probably very sensitive information that they will be supplying to you and the end client is a few tiers removed I'm very interested in that thank you so the question is how do we motivate suppliers to provide data it's a good question and it's gotten a lot easier in the last three years I can tell you that much suppliers are primarily motivated by commercial interests so they're motivated because brands in the US are requiring supply chain transparency in order to do business with with foreign suppliers and so having it now become part of CBP compliance is extremely motivating to overseas suppliers who increasingly have already mapped their supply chains and have the data available to share so like I said in the food and apparel and luxury industries the majority of tier one suppliers already have data available to share because this has been ongoing for more than a decade now there are a couple of other things at play here since many suppliers already provide transparency that provides a competitive advantage for those suppliers and a significant competitive disadvantage for companies that don't have the traceability yet so what has started to happen is a virtuous circle where suppliers are competing on the basis of being transparent and so as I showed you the slide with the different brands that are using supply chain transparency in a marketing way, in a brand building way suppliers are doing the same they're using supply chain transparency to sell more premium goods onto premium markets especially the North American ones so we think of supply chain transparency as a very small price to pay for access to the US market and then last but not least the biggest outlier group who are unable to provide transparency actually don't have the data or the technology and so what our outreach team does is identify anywhere that enabling technology needs to be offered and I forgot to say you know for suppliers to use source map is completely free so small, medium enterprises small holder farmers in West Africa Indonesia use source map every day because it's free and because it is a tool that they can use in and of itself to map their farms to upload their transactions and so we've worked very closely with our customers to provide that enabling technology and that could go we had recently a hazelnut farmer in Turkey that you know couldn't log in and our outreach team was able to track down the farmer's son and the son had an internet connection and was able to help his dad get online and upload the information so if the commercial incentive isn't enough more often than not it's the lack of access to technology and we have the experience to bridge that gap great thank you so we have a really kind of a bazing problem that I don't usually encounter which is that we're running a little bit early so what we're thinking is that instead of waiting and starting at 245 for the next session we'll take a quick break now and get started at 215 so let's come back here convene at 215 and we can let you go a little bit early today thanks a lot see you in a few minutes