Home Depot Builds Case Files for Police as Crime Syndicates Drive Shrink

犯罪集团加剧损耗，家得宝为警方建立案件卷宗 有组织盗窃团伙利用缺乏监管的点对点电商平台销赃，迫使家得宝为不堪重负的执法机构整理成套犯罪案卷，同时将易失窃工具上锁以保障销售。 Scott Glenn 深入剖析了家得宝（The Home Depot）零售损耗的现实情况，解释称约 70% 的损耗源自恶意盗窃和有组织犯罪团伙，而非简单的操作失误或偶发性顺手牵羊。讨论详细梳理了有组织零售犯罪的多层级链条——从招募扒手专门针对电动工具等高流动性商品下手的窃贼，到通过点对点在线平台洗白赃物的销赃网络——以及承运人冒用和货物盗窃等新兴威胁。Glenn 概述了正在部署的技术与运营应对措施，例如边缘 AI 视频分析、移动端二维码锁箱以及为执法部门整理的详尽案卷，同时权衡了 RFID 和工具“变砖”（远程锁死）等技术的财务取舍。主持人 Joe Weisenthal 和 Tracy Alloway 在总结中评估了防盗措施对顾客体验造成的额外阻碍，以及零售监控与类似 CORCA 等联邦立法行动在法律、技术和公民自由层面面临的挑战。 现代货物盗窃日益依赖承运人冒用手段，诈骗分子通过伪装成经核实认证的卡车公司，直接从发货装卸月台截走整车货物。 (0:00) 大型五金超市的被盗物品主要分为两类：诸如紧凑型电钻等高走量目标，以及随废品市场行情价格波动的铜线等高价值大宗商品目标。 (5:00) 家得宝将库存损耗分为两大主要类别：约 30% 为运营或管理差错，70% 为恶意盗窃。 (10:00) 疫情后线上二手交易市场的扩张，为被盗库存提供了高流动性且隐蔽去个人化的销赃渠道，从而极大助长了有组织零售犯罪。 (12:08) 鉴于地方警力面临严峻的资源短缺，企业资产保护团队在内部编制“完成度达 80%”的案件卷宗，以便警方能够迅速执行搜查与逮捕令。 (20:00)

Retail shrink at big-box chains has evolved beyond petty shoplifting into highly structured organized retail crime (ORC) syndicates that monetize stolen inventory through digital secondary markets and cargo exploits, according to Scott Glenn, Vice President of Asset Protection at The Home Depot. Anatomy of Retail Shrink and High-Theft Commodities (10:00) Retail shrink is divided internally at The Home Depot into operational or administrative losses ('things we do to ourselves') and malicious activity ('things people do to us'). Glenn reports that operational errors account for roughly 30% of shrink, while malicious theft represents the remaining 70%, with the bulk of losses concentrated across the chain's 2,400 physical stores rather than distribution hubs or online channels. Across the industry, baseline shrink was historically estimated at roughly 40% internal associate theft, 35% external shoplifting, and 20% to 25% administrative error, though variations in cost versus retail accounting methods complicate exact comparisons. Theft volume centers on compact, liquid products like power tools—particularly drills—while raw commodity price spikes make bulk electrical wire and copper the leading source of dollar-value loss. In extreme instances, Home Depot discovered it had purchased a specific $799 tool combination kit 4.2 times for every single unit actually sold. While securing items behind plexiglass introduces customer friction, Glenn notes that chronic out-of-stock conditions caused by theft can wipe out sales entirely, particularly among professional contractors who drive roughly 50% of the chain's revenue. The Syndicated Pipeline and Marketplace Loopholes (31:05) Organized retail crime functions as an illicit supply chain distinct from opportunistic shoplifting. Syndicates recruit vulnerable foot soldiers, known as 'boosters,' by providing upfront cash or drugs along with targeted shopping lists. Merchandise is transferred to 'fences' operating out of garages, warehouses, or flea markets, where retail labels and anti-theft tags are stripped. Resale profits are kicked up to supervisory figures ranging from local ringleaders to transnational gangs and cartels involved in narcotics distribution and money laundering. While the federal INFORM Consumers Act mandated identity verification for high-volume third-party sellers on major e-commerce platforms like Amazon and eBay, it largely left peer-to-peer marketplaces untouched. Consequently, fences migrated illicit inventory en masse to platforms like Facebook Marketplace and Craigslist. Fences prioritize liquidity over margin, listing unopened power tools at extreme discounts—such as selling a $499 drill kit for $200—while connected tech items generally retain 70% to 80% of retail value compared to 30% to 40% for everyday non-tech consumables. Glenn argues that peer-to-peer platforms resist self-policing due to compliance costs, forcing asset protection teams to scrape listings and conduct test buys themselves. Technical Bottlenecks and Upstream Cargo Vulnerabilities (23:45) Technological solutions face severe engineering and cost constraints inside large-format stores. Deploying comprehensive AI video analytics across a 140,000-square-foot warehouse requires cost-prohibitive server and cabling infrastructure. Similarly, passive RFID tags cannot reliably transmit through the structural steel racking and metallic merchandise of hardware stores, which function like a massive Faraday cage. Active RFID tags and manufacturer-level tool 'bricking'—embedding Bluetooth interlocks that prevent tools from operating until scanned at checkout—require expensive store sensor arrays and complex supply-chain re-engineering by tool manufacturers. Theft networks have also expanded upstream into logistics channels. Criminal rings exploit fragmented domestic trucking through carrier spoofing and AI-forged bills of lading, allowing fraudulent drivers to pick up full trailers from ports and transload yards. Cargo theft also includes physical break-ins while drivers sleep during mandatory rest stops, as well as targeted rail car raids where insider intelligence directs groups to breach specific containers carrying high-value electronics, connected tools, or firearms. Investigative Case Packaging and Legislative Momentum (20:00) To counter organized networks, Home Depot maintains a strict non-engagement policy to ensure associate safety, training 400,000 employees in de-escalation while treating frontline stops as intelligence-gathering opportunities. Because municipal police departments are constrained by violent crime caseloads and bail reforms, Home Depot's internal team of former federal and local law enforcement officers aggregates multi-store patterns and builds '80% complete' case files so authorities can secure search and arrest warrants with minimal friction. To accelerate this process, the chain installed upgraded video management systems across 2,300 stores to run edge analytics tracking spatial anomalies, aiming to compress the average 70-day investigative lifecycle. Because syndicates routinely move stolen inventory across state lines to exploit jurisdictional boundaries, the retail industry is backing the bipartisan Combating Organized Retail Crime Act (CORCA). Passed by the House and attached to the National Defense Authorization Act (NDAA), the bill establishes dedicated federal prosecutors and specialized coordination centers to align federal enforcement against retail, cargo, and payments fraud under interstate stolen property statutes. While advanced tracking and stricter legislative frameworks aim to compress the investigative timeline, retailers continue to balance financial loss against customer friction and civil liberties. As hosts Joe Weisenthal and Tracy Alloway observe, deploying behavioral edge analytics and app-based authenticated access to locked shelves risks transforming retail floors into high-surveillance identity checkpoints, all while physical enforcement remains ultimately dependent on overburdened municipal police departments.

彭博社《Odd Lots》节目指出，现代零售业商品损耗（Shrink）已从偶发性顺手牵羊演变为高度专业化、跨管辖区的有组织零售犯罪（ORC）；据The Home Depot资产保护副总裁Scott Glenn所述，零售商正被迫转型为准情报机构，以应对供应链盗窃、黑市转售及法律执法维度的系统性挑战。 损耗构成与有组织犯罪网络运作机制 (10:00) 零售损耗并不单纯源自货架被盗。Scott Glenn指出，在The Home Depot的损耗结构中，约30%属于内部操作或管理失误（即企业自身造成的损耗），其余约70%则为恶意盗窃，主要集中在全美2400家线下实体门店。在被盗商品中，电动工具因需求大且变现极快而成为盗窃数量最多的品类，而大宗铜料与电线则随原材料大宗商品价格波动，成为盗窃总金额最高的品类。早期的行业经验法则通常估计内部员工盗窃占40%、外部行窃占35%、管理失误占20%至25%，但随着近年来线上转售市场的发展，外部有组织盗窃的规模与危险程度被严重低估。 有组织零售犯罪（ORC）具备严密的分工链条：底层是被称为“扒手”（boosters）的执行人员，犯罪团伙通常以预付现金或毒品为诱饵与控制手段，向其分派明确的“扫货清单”；这些人员将物资运送至住宅车库、仓库或跳蚤市场的“收赃者”（fences）手中，后者机械化地清除防盗标签与门店条形码后，将赃物推向二级市场；获取的资金最终汇入更高层级的控制者手中，包括地方黑帮乃至跨国贩毒集团，用于洗钱、毒品分销及其他非法活动。 法案监管盲区与二级转售市场套现 (33:38) 盗窃物资的变现渠道近年来发生了重大转移。虽然美国于2023年生效的《告知消费者法案》（INFORM Consumers Act）要求大型电商平台核实并公开大额第三方卖家的真实身份，但该法规并未覆盖点对点（P2P）平台，促使大量非法转售迅速流向Facebook Marketplace和Craigslist等渠道。Glenn指出，由于缺乏严格的自律动力与法律强制要求，平台鲜少主动排查可疑卖家。在转售价格方面，高科技或联网工具以及礼品卡的黑市价格通常可达零售价的70%至80%，而除臭剂、泰诺等日常快消品往往仅以零售价的30%至40%出售以追求极速变现。 面对转售平台的监管缺失，The Home Depot等零售商不得不部署定制算法与网络爬虫来主动抓取挂牌信息。当发现类似“Tommy1234”等匿名账号以远低于批发成本的不可思议低价（例如将499美元的电钻套装挂牌200美元）成批出售原包装全新产品时，调查人员便以此为突破口展开“反向调查”，从数字店铺顺藤摸瓜锁定线下的收赃窝点及底层盗窃团伙。然而，在海量活跃广告中筛选证据，使零售商陷入了持续不断的被动应对状态。 实体门店防御考量与技术落地瓶颈 (23:45) 在实体门店内，将高价值商品锁入安全柜是资产保护部门与运营部门深度博弈的结果。Glenn以具体数据证实，在未加锁前，The Home Depot针对一款售价799美元的工具套装，每实际售出1套平均需采购4.2套，高损耗彻底摧毁了商品毛利；此外，专业承包商占公司总收入的约50%，若承包商因货架被掏空而无法在工期内拿到工具，将严重损害商业信誉。为此，门店内引入了QR码呼叫系统，将员工到场开锁的平均等待时间从传统呼叫器的最长10分钟缩短至约90秒，且加锁后部分货品因确保了正价库存充足，销售额反而出现增长。 然而，借助更前沿的技术全面杜绝损耗仍面临不可逾越的物理与经济限制。例如，将射频识别（RFID）技术全面应用于大型仓储式门店面临建筑钢梁、金属货架所形成的“法拉第笼”效应，无源标签信号会被严重屏蔽与反射，而高穿透性的有源RFID标签及全店读取硬件成本过于高昂。类似地，要在动辄14万平方英尺的仓库内全天候运行高密度AI视频行为分析，其服务器算力、布线与摄像头升级的软硬件投入（CapEx/OpEx）在财务上难以平衡；对于通过蓝牙或固件使工具在未付款前完全锁死（bricking）的设想，也受到工具制造商重组产线成本与全店传感器铺设费用的多重制约。 上游供应链漏洞与联邦立法协同推进 (45:00) 盗窃风险不仅发生于门店末端，更已大幅蔓延至整个物流上游。犯罪网络正利用AI高精度伪造提单（Bill of Lading）与提货凭证，直接冒充合法承运人从配送中心驶离整挂车的高价值物资；在货运长途运输中，团伙往往趁货车司机在法定休息时段睡眠期间撬锁洗劫；在港口及中转站，甚至出现集装箱被直接转运出海的现象；近五六年来，具备内线情报的犯罪团伙更是频频袭击铁路货场停滞车厢，精准破拆装有昂贵电子产品与工具的特定箱体。 由于地方警力资源紧张、优先处置暴力重案，加之保释金改革导致轻罪人员快速释放，零售商无法指望警方自主侦办零售窃案。The Home Depot组建了由前执法人员构成的专业团队，独立完成包含轨迹分析与收赃追踪的“完成度达80%”的案件卷宗，将调查周期从平均70天进一步压缩，以便检方直接向法官申请搜查令。在立法层面，零售行业正积极推动跨党派的《打击有组织零售犯罪法案》（CORCA），依托跨州盗窃财产罪名（18 U.S.C. § 2314）打破地方行政壁垒，设立专职联邦检察官与协调中心，从联邦层级系统性遏制跨州犯罪网络。 当前零售商通过边缘视频分析、货架数字身份验证以及深度卷宗调查来反制盗窃，但员工安全第一的“不介入”准则与警力追责瓶颈，使得现场抓捕依然受限。随着立法推进与技术升级，零售防御逐渐演变为涵盖供应链反欺诈、平台合规审查与跨州司法协同的长期系统工程，如何在保护资产的同时避免顾客体验过度摩擦化与公共隐私侵蚀，仍是悬而未决的核心议题。

## Background 本期节目由彭博社（Bloomberg）于 2026 年 9 月 10 日发布在播客平台，时长约 52 分钟。节目简介指出，在纽约等城市的药店，大量日常商品因防盗被锁在玻璃柜中，本次访谈邀请家得宝（The Home Depot）资产保护副总裁 Scott Glenn 深入探讨零售损耗的严峻现状、有组织零售犯罪（ORC）背后的黑市网络，以及零售巨头如何平衡顾客体验与库存保护。 ## TL;DR 节目全面揭开了全美零售业面临的体系化“有组织零售犯罪”（ORC）内幕。家得宝防损负责人 Scott Glenn 详述了盗窃光谱的两极——从底层的机会型顺手牵羊，到牵涉跨国黑帮、通过现金与毒品操控底层“扒手”的职业犯罪网络，并指出电钻等电动工具和铜缆是损失最高的重灾区。面对警力短缺、免现金保释等现实困境，零售商主动充当侦查先锋，制作高度完善的案卷移交警方，同时通过边缘视觉 AI、全美 2,300 家门店系统升级等技术手段压制犯罪。面对货架普遍“上锁”引发的体验争议，Glenn 算清了财务账本（部分高价值商品每卖 1 件需补货 4.2 次），证明物理锁具反而在一定程度上保住了真实货架库存，同时节目还探讨了货运供应链截流、二手平台监管漏洞及《打击有组织零售犯罪法案》（CORCA）的立法前景。 ## Atmosphere & style 整期对话专业严谨且兼具一线的硬核刑侦纪实感，两位主持人 Tracy Alloway 和 Joe Weisenthal 配合默契，以好奇且充满质疑精神的提问不断推敲商业防盗逻辑。Scott Glenn 身着标志性橙色围裙连线，以法学与刑侦专业背景出身的沉稳语调娓娓道来，既有坦陈“员工曾被击伤乃至遇害”时凝重冷峻的人性关怀，也有拆解“法拉第笼”物理困境与黑市定价心理时的幽默与务实。尾声主持人的复盘更交织着一丝黑色幽默与科技反乌托邦色彩，生动呈现出零售业在“打地鼠”式防盗博弈中的疲惫与清醒。 ## Bulletin - `06:40` 有组织零售犯罪（ORC）已演变为高度体系化的跨国重罪网络，非法所得常被用于资助毒品走私、洗钱及人口贩卖，远超偶发性顺手牵羊的范畴。 - `42:35` 家得宝数据显示极端的货损财务账本：为达成 1 件高端工具套件的合法销售，门店平均需进货 4.2 次，意味着超过 3 件在销售前即遭盗窃。 - `44:59` 物理加锁防盗在多品类中反倒提振了销售额，原因在于加锁前货架常年被盗空呈现“假性缺货”，加锁维持了物理库存反而释放了真实购买需求。 - `52:15` 美众议院推动的《打击有组织零售犯罪法案》（CORCA）已附加至 NDAA，拟设立国家级统筹中心，将跨州销赃与供应链劫掠正式纳入联邦执法重点。 ## Key takeaways - `00:00` 现代物流与零售盗窃呈现高度体系化，从伪造承运商身份整车拉走货物到铁路死轨精准撬车，供应链盗窃技术与情报水准大幅升级。 - `04:55` 按被盗件数计，手电钻等电动工具高居榜首；若按损失金额计，随行就市且极易变现的铜线与金属原材料是家得宝被盗最多的品类。 - `10:00` 零售损耗被划分为约 30% 的可控运营管理损耗与 70% 的恶意损耗，后者跨越供应链、实体门店及电商全渠道，尤以实体店最为严重。 - `12:30` 疫情冲击与线上全渠道转售通道的爆炸式扩张，为犯罪分子提供了去人脸化的洗赃黑市，导致官方统计严重滞后并低估了 ORC 的真实破坏力。 - `21:37` 针对警力严重短缺的现状，零售商防损团队多由前联邦探员主导，负责完成 80% 的前期取证卷宗，以便警方与法官能直接高效签发搜查与逮捕令。 - `21:37` 外界盛传零售商‘放任小偷以凑够重罪金额’纯属误解，防损人员暂缓收网通常是为了跟踪外围路线，连根端掉二级销赃窝点（Fence）。 - `24:00` 14 万平方英尺的超大仓储式卖场使全店 AI 视觉无死角追踪面临极其高昂的算力与硬件成本，企业必须在损耗承受度与投资回报间权衡。 - `26:40` 由于犯罪集团多雇佣身患毒瘾或居无定所的人员行窃，极易引发致命冲突，家得宝坚持‘用货损换安全’，严禁普通员工进行暴力阻拦。 - `28:45` 仓储式卖场密集的钢制货架形成天然的‘法拉第笼’，导致廉价无源 RFID 信号严重衰减失效，而高成本的有源方案在工程经济上难以全面铺开。 - `30:00` ORC 链条分工严密：由一线扒手负责盗窃，销赃中介（Fence）负责清洗标签及条码，最终所得资金层层抽成回流至顶层犯罪首脑。 - `32:53` 《消费者知情法案》（INFORM Act）促使主流电商强化卖家实名审核，但犯罪团伙迅速转向缺乏严格实名义务的 P2P 平台（如 Facebook Marketplace）。 - `40:00` 黑市定价呈现极高‘市场效率’：科技含量高且可联网的高端工具能以正品 70%~80% 变现，而日用易耗品通常只能卖到 30%~40%。 - `46:17` 干线物流与铁路车厢盗窃呈现高度情报化渗透，盗贼能精准避开普通建材直奔装载枪支或精密设备的特定车厢，表明内部人员极可能被收买泄漏清单。 - `50:00` 家得宝通过全面升级 2,300 家门店的边缘视频分析系统与线上网络爬虫，全力将 ORC 案件从侦测到移交警方的调查闭环周期压缩至 70 天以内。 - `56:15` 防盗治理面临‘打地鼠’困局：技术拦截与货架上锁不仅带来隐私争议与准入摩擦，更驱使犯罪网络向干线劫持、虚假提单及退货掉包欺诈等新形态转移。 ## Detailed walkthrough ### `00:00` 货运盗窃新形态与被锁住的零售货架 在现代零售与物流供应链中，盗窃的方式正在经历意想不到的剧变。播客主持人 Tracy Alloway 和 Joe Weisenthal 谈及此前与 Truck Parking Club 首席营销官 Reid Lustig 的一次交流，那期节目揭示了美国货运反弹背景下愈演愈烈的“货物盗窃”（cargo theft）危机。在普通人的想象中，货运盗窃或许只是一伙毛贼趁夜从卡车车厢后偷拽几个箱子；但现实中的手段已经演变为高度体系化的欺诈：犯罪分子直接伪造或冒充合法的承运商身份（carrier spoofing），大摇大摆地开进物流节点拉走整车货物，转头在黑市脱手。全美有数万家小型货运公司，托运人与承运商之间的核验往往存在漏洞，让这种“凭空冒出并卷走整车货物”的骗局屡见不鲜。 与供应链深处的暗流相呼应，城市日常零售终端的安全危机也愈发刺眼。《彭博社》近期报道了现代版“火车大劫案”（大量铁路线货运车厢遭到沿途洗劫），而在纽约等大城市的街头，实体零售店的防盗措施更是直接打破了消费者的日常体验。只要走进街角的 Duane Reade 或各类药妆店，就会发现货架上几乎所有日用品都被锁在透明玻璃柜后。Joe 指出，不论外界如何争论疫情后入店行窃的统计数据是否被夸大，零售商把除臭剂、洗发水等低价日常品统统锁死是不可否认的现实。品牌方和零售巨头心知肚明这会严重破坏购物体验，也清楚需要店员拿着钥匙随时待命开锁会拉低人效，但他们依然不得不选择牺牲客流体验，这足以说明背后的盗窃损失已经严重到了何种程度。 --- ### `04:55` 家得宝防损掌舵人登场：最常被偷的究竟是什么？ 为了厘清从货架上锁到跨国物流劫案的完整链条，节目请来了家得宝（The Home Depot）资产保护副总裁 Scott Glenn。Scott 当天特意身着家得宝标志性的橙色围裙连线。他笑着解释，在财报发布日这不仅是支持公司，更是企业文化的核心载体——工装围裙上“将价值观置于胸膛左侧靠近心脏的位置”是家得宝独特的传统设计。 谈及大家最关心的八卦问题——家得宝被偷得最多的商品到底是什么？Scott 将答案拆分为“件数数量”（Volume）和“涉案金额”（Dollars）两个维度： ```mermaid flowchart TD accTitle: 家得宝高发被盗商品分类 accDescr: 展示按被盗件数与按被盗总金额划分的两大家得宝高发失窃商品类型及特征。 subgraph Volume["按被盗件数排行 (Volume)"] A["电动工具（如手电钻）"] --> B["转售变现快、体积适中、需求普遍"] end subgraph Dollars["按涉案金额排行 (Dollars)"] C["铜线与金属原材料 (Copper & Wire)"] --> D["大宗商品属性、价格随行就市、黑市销路稳定"] end ``` 从被盗件数来看，手电钻等电动工具常年高居榜首；而如果按损失金额计算，被盗最多的则是铜与电线线缆。铜作为大宗商品具有极强的变现能力，犯罪分子一旦得手，既能在二手黑市转卖，也能直接用于非法承揽的工程项目。Tracy 补充了一个身边的真实案例：康涅狄格州建于殖民时期的内森·黑尔博物馆（Nathan Hale Museum）前不久遭人破门而入，窃贼竟然硬生生拆走了建筑内所有珍贵的老旧铜水管。Scott 对此感同身受地表示，八年前他刚入职家得宝并在当地购房，在房产过户交割与正式搬入之间的短暂空档期，小偷就潜入他的新家，将暖通空调（HVAC）外机及管道里的铜管全部洗劫一空。铜价具有周期性波动的特点，一旦国际铜价飙升，这类金属盗窃案便会呈现爆发式增长。 --- ### `06:40` 从抓小偷到对抗跨国集团：零售盗窃的完整光谱 回顾自己的职业生涯，Scott 坦言自己走到今天纯属意外。他大学原本攻读建筑学，后来发现自己数学天赋有限，加之一直对法律和司法抱有极大兴趣，便转而考虑加入 FBI 并攻读了法学院。法学院毕业后，他曾进入美国财政部工作，但体制内的稳定并非其长久志向。幸运的是，大学暑期他在 Target 百货做过抓店内小偷的兼职安保；离开政府部门后，凭借这段早期防损经验，他重新跨入零售行业，一步步在资产保护领域晋升，先后在 Kohl's 百货工作十年、在 Sears 担任了十年的首席安全官（CSO），直至八年前加入家得宝掌舵安全防损业务。 针对公众常说的“零售盗窃”，Scott 详细拆解了内部所划分的“盗窃光谱”： * **低端/机会型盗窃（Opportunistic & Need-based）：** 光谱的一端是低阶的入店顺手牵羊或迫于生计的盗窃。例如生活陷入困境、无力承担日常开销的个体，或者是顾客在自助结账机前无心或轻微贪小便宜少扫了一支除臭剂。对于这类行为，家得宝虽不认可，但这绝非防损团队的核心攻防重心，也不是投入海量技术与资源的领域。 * **高端/有组织零售犯罪（Organized Retail Crime, ORC）：** 光谱的另一端则是高度职业化、网络化的庞大犯罪集团。这些团伙背后牵扯到跨国犯罪网络、黑帮势力甚至毒品卡车集团（cartel-related, transnational, gang-related organizations）。他们盗窃的目的绝非为了自用，而是将大规模盗窃作为变现工具，用赃款来为洗钱、毒品走私和人口贩卖等极其严重的跨国重罪提供运作资金。 * **攻防焦点的转移：** 家得宝的防损资源、情报系统和战术策略，绝大部分都集中压在这一类处于金字塔顶端、对企业和社会治安造成毁灭性破坏的组织化犯罪上。 随着讨论的深入，话题进而引向了财报和零售业中最常提及的专业术语——“损耗/缩水”（Shrinkage）。盗窃如何被算入这笔账、在供应链与门店之间究竟扮演了怎样的角色，成为了下一阶段探讨的核心。 ### `10:00` 零售损耗的本质与分类：“自身原因”与“外部侵害” 零售行业所说的“损耗”（shrink），最底层的逻辑非常简单：账面上有这些库存，但到了实际盘点核算时，东西凭空消失了。然而，这种消失绝不仅仅是“坏人冲进店里偷东西”那么单向度。家得宝（The Home Depot）资产保护副总裁 Scott Glenn 向管理层和董事会汇报时，通常将损耗清晰地划分为两大维度：一类是“运营性或可控管理型损耗”（operational or controllable administrative shrink），另一类则是“恶意损耗”（malicious shrink）。用大白话讲，就是“我们自己造成的”（things we do to ourself）与“别人对我们施加的”（things people do to us）。 在家得宝内部，得益于扎实精细的管理体系，可控的运营管理损耗大约占到总体损耗的 30%，其余约 70% 则属于外来侵吞或恶意盗窃。从物理链条来看，损耗被拆分在三个相对独立的环节中考量并最终汇总：供应链体系（分拣中心、大型仓库及运输干线）、实体门店网络，以及线上电商渠道。在上述任何一个节点中，都同时存在着运营管理错误、财务记账偏差与恶意侵占。但毫无疑问，在拥有约 2,400 家大型门店的家得宝网络中，损耗最集中、规模最大的板块依然发生在实体门店现场。 ```mermaid flowchart TD accTitle: 零售总损耗归集结构 accDescr: 展示供应链、实体店和线上渠道各自的运营与恶意损耗如何汇入总损耗 subgraph Channels["损耗发生场景"] SC["供应链 / 仓储运输"] ST["2,400 家实体门店 (占比最大)"] ON["线上电商渠道"] end subgraph ShrinkTypes["损耗属性拆分"] OP["运营/行政损耗 (约 30%)\n如记账错误、内部自损"] MAL["恶意损耗 (约 70%)\n如内部盗窃、有组织偷盗"] end SC --> OP SC --> MAL ST --> OP ST --> MAL ON --> OP ON --> MAL OP --> Total["公司总体零售损耗 (数百亿美元行业规模)"] MAL --> Total ``` --- ### `12:30` 有组织零售犯罪（ORC）的演变与数据失真争议 当媒体大肆渲染“Target 因犯罪率飙升关闭门店”或各零售巨头损耗激增时，行业内部实际上深陷一场巨大的统计与定义争议。损耗规模全行业早已达到数百亿甚至数百亿美元，但各家公司的核算基准大相径庭：家得宝采用的是零售核算制（retail accounting），而另一些零售商则采用成本核算制（cost accounting），加上加价率（markup）后数额差异极大。在传统的行业经验法则中，恶意损耗往往被粗略拆解为：40% 来自员工内鬼盗窃，35% 来自外部人员入店行窃，剩下 20% 到 25% 归因于内部运营疏漏。 然而，Scott Glenn 坦言，当他于 2018 年刚加入家得宝时，他甚至根本不相信所谓“有组织零售犯罪”（Organized Retail Crime, ORC）是一个真实且严峻的独立命题。但随后几年的剧烈变化彻底颠覆了他的认知：一方面是新冠疫情的冲击；另一方面是“全渠道”（Omnichannel）销售网络的爆炸式扩张，为不法分子提供了能够极为匿名、去中心化销赃的线上黑市通道。现如今，ORC 不仅利润越来越丰厚、渗透无孔不入，而且其暴力和对抗程度急剧上升，直接危及店内员工与顾客的人身安全。更为关键的是，由于疫情后报案率下降与统计机制滞后，无论是行业协会（如 RILA、NRF）的估算还是 FBI 的全美统一犯罪报告（UCR），都严重低估并低报了这一问题的真正严重性。 - 行业传统损失比例估计：内部员工盗窃占约 40%，外部顾客行窃占约 35%，运营管理自损占 20%~25%。 - ORC 特性演变：线上转售（Omni-resale）渠道让洗赃过程完全去人脸化，导致犯罪呈现链条化与高暴利。 - 数据盲区：官方警情通报与犯罪数据库中记录的数字，远远落后于现实门店发生的频繁盗窃。 --- ### `18:40` 从普通顺手牵羊到有组织劫掠：侦查视角的切入点 区分偶发性的个体行窃与严密分工的“有组织零售犯罪”，是资产保护一线面临的最大难题。在单起盗窃行为发生的当下，二者在肉眼观察上几乎毫无区别——有人走近货架，抓起工具塞进衣服或推车，然后堂而皇之地走出店门。 家得宝的安全防护力量采取“分级联动”模式：一线的驻店安保人员专职负责现场抓现行、控制嫌疑人并移交属地警方；而在亚特兰大总部，则常驻着专门针对跨区域、高精尖犯罪网络的情报调查团队。现场的审讯与细节问询往往是撕开面具的关键切口。安保人员把偷了一把电钻的嫌疑人拦下并询问时，如果对方老实交代只是因为手头拮据、想修缮自家后院的木甲板（deck），这属于典型的孤立个人盗窃；但如果同一件工具背后牵扯出固定收购点、反复踩点路线或特定的商品品类需求，整起事件就会立即上升为总部级 ORC 追踪的线索起点。 ### `20:00` 顺藤摸瓜：从底层“扒手”拼凑有组织犯罪网络 在防损调查的一线，掌握犯罪网络全貌的关键往往始于对现场扒手（Booster）的讯问。当家得宝（The Home Depot）的防损人员截获底层行窃者时，经常会听到极其类似的供词：“有人让我进店偷这个，之后带到另一个地方去交差；我可以从中换取现金或毒品，他们还会给我下一张清单，让我去另一家分店继续偷。”防损团队系统性地收集这些一线口供与情报，并将其汇总给总部的有组织零售犯罪（ORC）调查专家。这些专家多为联邦调查局（FBI）、州或地方执法机构的退役探员，擅长从海量繁杂的信息中“去伪存真”，甄别出哪些线索具备深挖价值。 现行的司法困境使得跨店案件的串并显得尤为必要。在许多地区，诸如“持有特定驾照的嫌疑人约翰尼”可能在短短时间内连续光顾四家门店行窃，即便被抓，由于保释金改革或免现金保释（no-cash bail）政策，他们往往隔天就能被释放，随即重操旧业。因此，家得宝的专家必须通过数据分析将散落在不同门店的零碎报案串联起来，判断哪些是值得动用大量资源深挖的职业犯罪团伙，哪些纯属偶发性的一般小偷小摸。 ```mermaid flowchart TD accTitle: 零售犯罪调查与执法协同流程 accDescr: 展示从一线扒手口供到防损情报分析，再到向警方移交80%案件卷宗的完整链条 subgraph StoreLevel ["门店一线收集"] A["截获底层行窃者（扒手）"] --> B["获取关键口供与作案清单"] B --> C["汇总跨门店活动记录"] end subgraph AnalysisLevel ["总部情报研判"] C --> D["ORC专家团队（前执法人员）"] D --> E["去伪存真：筛除单次偶然作案"] D --> F["串并案件：锁定职业销赃团伙"] end subgraph LegalLevel ["司法打击落地"] F --> G["制作完成度达80%的成熟卷宗"] G --> H["警方/法官迅速申请搜查与逮捕令"] end ``` - 调查团队核心构成：大量吸纳来自前联邦探员、州警及地方执法部门的资深刑侦专家。 - 司法现实瓶颈：免现金保释等政策导致底层窃贼“前脚进局子、后脚回街头”，必须依靠团伙案件串并才能形成有效威慑。 --- ### `21:37` 替警力减负与破除“金额达标才抓人”的迷思 主持人不禁质疑：顺藤摸瓜调查整个犯罪链条，难道不该是警察的工作吗？斯科特·格伦（Scott Glenn）坦言，这绝非指责警方，而是残酷的资源现实所致。如今各地警察部门普遍警力吃紧、疲于奔命。在执法的优先等级序列中，谋杀、武装劫持、劫车等恶性暴力犯罪永远排在零售盗窃前面，警方根本无暇抽身主动替商场侦办复杂的跨区域盗窃案。因此，家得宝的内部策略是“替警察把前期的苦活累活干完”，直接向警方移交一份“完成度达到80%”的详尽案件卷宗，让警察只需拿着材料走向法官宣誓并申请搜查令与逮捕令。这种高度准备好的案件，执法机关接手的意愿往往远高于毫无线绪的普通报案。 面对外界长期流传的谣言——“零售巨头故意放任小偷行窃，直到盗窃金额累计达到重罪门槛才收网抓人”，格伦给予了坚决否认。他指出这种说法视角过于狭隘，外界看到的“眼睁睁看贼走人”，实际上往往是针对惯犯的长期收网行动之一。比如调查人员早已掌握某人过去30天内席卷了50家门店，此时放其离开并非为了“凑金额”，而是安排外围跟踪，顺藤摸瓜去端掉背后的销赃据点（Fence）、隐秘仓库或分销网络。家得宝追求的是连根拔起整个犯罪团伙，而非纠缠于单次盗窃的金额数字。 - 警民协同法则：企业主动承担繁重的前期取证工作，制作现成卷宗以适应警力匮乏的司法现实。 - 辟谣重罪门槛策略：不以“凑够起诉金额”为目的，允许涉案人员出门多为追踪二级销赃窝点或货运路线的战术需要。 --- ### `24:00` 技术防治的现实瓶颈：14万平方英尺仓库的算力与成本博弈 既然顺手牵羊拿走一个小垫圈难以防范，但像电动工具这样的大件商品，为何不能单纯靠自动化技术彻底解决？主持人设想了一种技术闭环：给每件工具贴上防盗标签，天花板配备全覆盖摄像头，结合AI视觉算法捕捉顾客从货架取下电钻、绕过收银台直奔大门的行为，从而自动拉响警报。格伦对此幽默地回应道：“我很欣赏你的思路，但那真是一个非常昂贵（expensive）的思路。” 家得宝的单店规模普遍达140,000平方英尺（约合1.3万平方米），且本质上是一个高密度的“作业式仓库”。要在如此广阔的立体空间内实现全程无盲区的高级AI轨迹追踪，所需的摄像头数量、硬件设备以及后端的边缘计算算力支出将是天文数字。尽管家得宝在部分重点通道或前端收银区部署了前沿的AI视频分析系统，但全店铺设涉及巨大的资本支出（CapEx）与运营成本（OpEx）。零售企业必须在承受损耗与技术投资之间做出权衡——“虽然我们渴望将损耗清零，但最终能达到的现实数字总会高于零”。 针对产品“变砖（Bricking）”技术（即未付款商品无法激活使用），家得宝也曾深度尝试。例如联合供应商在 Milwaukee（美沃奇）电钻等高端工具内部植入蓝牙芯片，一旦未经收银台激活就彻底锁死。然而，这种方案需要重构制造商的整条生产线，而且门店仍然必须部署昂贵的基站硬件来感应与激活芯片，导致整体方案极难在全品类中大规模推广。 - 空间与算力挑战：140,000平方英尺的超大卖场使得高密度AI视觉追踪的软硬件铺设成本畸高。 - “变砖”防盗试验：曾探索在高端电动工具中植入蓝牙锁止芯片，但因供应商产线改造成本及门店基站开销过大而难以全面铺开。 --- ### `26:40` 人身安全重于一切：有组织犯罪背后的暴力威胁与员工守则 比起财务账面上的货损，更令管理层神经紧绷的是现场的人身安全隐患。在社交媒体上，常有公众对“窃贼推着塞满商品的购物车大摇大摆走出大门，保安与店员却袖手旁观”的画面表示不解。格伦揭示了其背后残酷的治安危机：有组织犯罪网络极度狡黠，他们常常招募并利用流浪者、严重药物成瘾者等社会边缘群体充当一线扒手。这些人潜入商场的唯一目标就是“必须把货带出去换取毒资或微薄赏金”，在此极端动机驱使下，他们极易被激怒并铤而走险采取暴力行为。 暴力对抗在家得宝并非假设，而是发生过的血淋淋现实。格伦神情凝重地指出，暴力冲突不仅波及整个零售业，在家得宝门店内也真实发生过店员“被击伤甚至被杀害”的惨剧。基于此，公司的安全铁律被置于任何财产之上。尽管店内400,000名员工对工作充满热情、极其痛恨偷窃，但公司对所有员工及驻店防损专员进行严格的降级冲突（De-escalation）培训，严令禁止在嫌疑人逃跑时进行身体阻拦或暴力对抗。格伦斩钉截铁地强调：“我宁愿每天用货损换取人员的安全，公司的执行领导层和董事会百分之百支持这一决定——我们绝不会把员工和顾客置于危险之中。” - 一线行窃人员特征：犯罪集团多雇佣身患毒瘾或居无定所的人员行窃，极具绝望感和暴力倾向。 - 铁律底线：家得宝严禁普通员工介入盗窃阻拦；400,000名员工均接受冲突降级与回避培训。 - 核心管理原则：“用货损换安全”，在极端暴力威胁面前，人身生命安全永远享有最高优先级。 --- ### `28:45` RFID 为何未成灵丹妙药：钢架结构下的“法拉第笼”困境 针对防盗标签应用了25年为何仍未能普及且成本高昂的疑问，格伦深有共鸣地苦笑称这真是“把大家的心愿说给了上帝听（your lips to God's ears）”。从零售管理的角度来看，RFID不仅关乎防盗，在商品全渠道追踪、库存可视化、货架自动补货等供应链环节具有无可比拟的商业价值。然而，家得宝独特的仓储卖场环境对无线射频技术提出了近乎严苛的物理挑战。 首先，家得宝卖场内部搭建着漫无边际的重型工业级钢制货架，这些密集的金属结构在物理层面上直接形成了一个巨大的“法拉第笼（Faraday cage）”，对无线电信号造成严重屏蔽与多径反射。这导致在常规服装店表现优异的低成本无源RFID（Passive RFID）在家得宝基本失效。若要实现稳定感应，必须采用能够主动发射信号的有源微型射频芯片（Active pinging RFID），而这类芯片与配套读取装置的硬件造价极为高昂。再加上每家门店通常包含多达8个各自分立的进出大门（包括大件建材出入口、园艺区出口等），要把每一个门禁节点都武装上高精度的抗干扰读取硬件，在现阶段仍是一道难以克服的经济与工程算术题。 - 物理屏蔽难题：仓储式货架形成天然的金属“法拉第笼”，廉价的无源RFID信号严重衰减无法正常工作。 - 硬件落地障碍：门店必须采用高昂的有源发射RFID技术，叠加全店多达8处复杂出入口，导致技术全面落地成本远超损耗本身。 ### `30:00` RFID 的物理局限与“销赃网络”（Fencing）的运作机制 在讨论实体防盗技术时，Scott Glenn 澄清了为什么 RFID（射频识别技术）很难直接在像家得宝（The Home Depot）这样的仓储式大卖场中普及。关键障碍并非电子标签本身的成本——如今被动式 RFID 标签的价格已经低到微不足道——而是店内的物理结构。大卖场内部充斥着密集的钢制货架，这些钢铁材料会直接阻断无线射频信号。虽然市场上存在信号更强的主动式（有源）RFID 技术，但在整个门店内大范围布置所需的高规格读取硬件依然极其昂贵。 话题随后转向商品离开零售终端后的黑色产业链：销赃网络（Fencing）。Glenn 深入拆解了有组织零售犯罪（ORC）的分工架构： - **行窃者（Boosters）**：他们是处在最前线承担被捕风险的人。他们通常由犯罪集团招募，领着特定的“盗窃清单”，甚至会提前收到现金预付款。在很多情况下，幕后黑手会用毒品或“交货换毒品”的承诺来控制他们，防止他们拿钱跑路。这些人在门店内作案时目标极其明确、不择手段，甚至在撤离时表现得极为暴力和激进。 - **销赃中介/收赃人（The Fence）**：Boosters 得手后，会将整批货物运送到中介处。这个节点可能是某个民宅车库、租赁仓库、跳蚤市场等各种隐秘地点。Fence 的核心工作是对货物进行“清洗”（cleaning）——撕掉 Home Depot、Target、CVS 等零售商的价格标签与条形码，拆除防盗磁扣或电子防盗标签，掩盖货物的非法来源。 - **顶层犯罪首脑（The Bosses）**：销赃所得的资金会按比例一级级向上层“抽成”。幕后操盘者背景跨度极大，可能只是几个合伙租房搞犯罪发家的小团伙，也可能是跨国毒枭卡特尔、南美跨国盗窃集团或来自其他地区的跨国犯罪网络。 ```mermaid flowchart TD accTitle: 组织化零售犯罪运作链条 accDescr: 展现从顶层犯罪首脑到底层行窃者、销赃中介以及二手平台的分工链路 subgraph Command ["上层指挥与控制"] Boss["犯罪集团首脑 / 卡特尔 / 跨国犯罪团伙"] end subgraph Frontline ["一线盗窃执行"] Booster["底层行窃者 (Boosters)<br/>拿订单清单、现金或毒品控制"] Store["零售大卖场<br/>(Home Depot / CVS / Target)"] end subgraph FencingChain ["销赃流转与洗白"] Fence["销赃中介 (Fence)<br/>车库、仓库、跳蚤市场"] Cleaning["商品清洗<br/>拆防盗扣、撕标签、掩盖来源"] end subgraph Resale ["转售与套现"] Marketplaces["P2P 转售平台<br/>(Facebook Marketplace / Craigslist)"] EndBuyer["不知情的终端买家"] end Boss -- "下达偷窃清单与资金控制" --> Booster Booster -- "暴力/团伙盗窃" --> Store Store -- "赃物出店" --> Booster Booster -- "送交货物" --> Fence Fence --> Cleaning Cleaning --> Marketplaces Marketplaces --> EndBuyer EndBuyer -- "货款回流" --> Fence Fence -- "洗钱与利润抽成上缴" --> Boss ``` ### `32:53` 平台监管漏洞与《消费者知情法案》（INFORM Act） 赃物清洗完毕后的转售渠道，在近几年发生了翻天覆地的转变。在 2010 年代末至 2020 年代初，大量盗窃物品是直接通过 Amazon、eBay 等主流大型电商平台倒卖的。然而，随着 2023 年联邦《消费者知情法案》（INFORM Consumers Act）的生效，大型第三方电商平台被强制要求核实大额卖家的真实身份、银行账户及税务信息（即落实“了解你的卖家”审查），这使得主流平台上的赃物流通大幅受阻。 然而，犯罪团伙迅速转向了**点对点（P2P）平台**，例如 Facebook Marketplace 和 Craigslist。由于法案立法时存在豁免漏洞，P2P 交易平台并未被纳入与大型电商平台同等严格的商户身份审查框架中。 - **监管成本差异**：审查平台上海量散户的真实身份和转售资质需要极其高昂的审核成本。在法律没有强制施压的情况下，P2P 平台缺乏主动投入资源建立严格治理机制的意愿。 - **家得宝的逆向侦查**：由于平台治理不足，家得宝内部的安全技术团队不得不建立监控模型，主动巡视各大线上二手平台。通过数据算法捕捉异常交易——例如分析为何网名为“Tommy1234”的用户能以 200 美元的超低价格长期、大批量售卖原价 499 美元且“全新未拆封”（New in Box）的组合电钻套件，该售价甚至远低于家得宝的进货成本。这种线上数据线索往往成为资产保护团队“逆向倒查”实体店行窃者的起点。 ### `36:34` 赞助插播：商业动态与科技工具 - **彭博科技快讯（Bloomberg Tech Minute）**：Carol Masser 报道了一项针对全球商学院雇主的调查。尽管企业普遍认同 AI 是未来的发展方向，但雇主认为当前商学院毕业生对 AI 工具的实际应用能力准备不足。在现阶段的招聘中，沟通能力、解决问题能力和适应力依然排名前列，AI 技能目前主要被期望用于日常流程的自动化处理。 - **赞助商公告**：节目期间穿插了辛辛那提保险公司（Cincinnati Insurance）关于企业风险防范的代理服务介绍，以及协作平台 Asana 针对人机团队与 AI 智能体工作流（AI Teammates）协作系统的商业推广。 ### `39:40` 销赃黑市的定价策略之谜 在节目尾声，主持人提出一个耐人寻味的定价博弈问题：为什么销赃中介在转售这些全新未拆封的高价值工具时，倾向于选择极大幅度的打折跳楼价（例如以两折、三折贱卖），而不是挂出接近原价的“微小折扣”以换取更高的利润率？他们似乎主动将利润留在了桌上。Scott Glenn 坦言，虽然自己并未对黑市销赃者的定价心理学做过学术级的深入剖析，但快速清空库存套现、避免持有大量赃物带来的法律与仓储风险，显然是这套黑色周转逻辑中的核心考量。 ### `40:00` 黑市定价的高效市场与平台监管的缺失 黑市对被盗物品的定价机制体现出惊人的“市场效率”：商品的科技含量越高、联网能力越强，其在二级市场的售价就越贴近正品零售价。礼品卡与具备智能互联功能的高端电动工具，往往能以零售价 70% 到 80% 的高位在黑市快速变现；相比之下，除臭剂、T恤衫或泰诺（Tylenol）这类日常易耗品，转手价格通常仅为零售标价的 30% 到 40%。犯罪团伙非常清楚哪些高单价、高流动性的商品值得冒险成规模地成批盗掘。 针对二手交易平台存在的监管漏洞，话题转向了 Meta 等巨头在治理被盗商品流通方面的被动态度。尽管 Meta 设有专门的安全审核部门并与零售商保持沟通，并非完全视而不见，但其实际作为与其实力严重不符。目前整个流程是“本末倒置”的：往往是 Home Depot 的资产保护团队通过实地暗访、自行掏钱购买赃物、逆向追踪到货品出自哪家具体门店、固定其犯罪证据后，再去找平台交涉，平台才会下架违规账号。这让零售商感到既失望又费解。平台明明掌握着底层交易数据，只要系统发现某个个人卖家在以远低于 Home Depot 或 Target 的价格大批量、规模化倾销全新工业级工具，这本身就该触发风控红线将其封禁。平台理应建立常态化的商户身份核验、税务申报追踪（如类似 1099-K 表格的合规机制），承担起与 Amazon 和 eBay 同等的法定义务，而不是依赖受害企业上门当“吹哨人”。 ### `42:35` 防盗加锁背后的财务账本：每卖 1 件要进 4.2 次货 在实体零售店中，将商品关进塑料防盗柜并要求顾客按铃等待，常常给消费者带来极度糟糕的购物体验——顾客经常按了铃苦等十分钟，等来的店员却发现自己根本没拿对应的钥匙。但在零售巨头内部，把商品锁起来绝非拍脑袋的决定。这是一场资产保护团队与采购商（merchants）、门店运营团队（operators）之间极其艰难且火药味十足的博弈。2018 年年中 Scott Glenn 履新时，Home Depot 的货损（shrink）正处于持续恶化的危险轨道上。为了说服运营部门，他用显微镜般精细的数据做了一次财务穿透分析。 团队调取了一款售价 799 美元的顶级工具组合套件的数据，得出的结论让财务部门大为震惊：**为了达成 1 次真实销售，门店平均需要向供应商采购 4.2 次该商品**。这意味着有超过 3 件在销售之前就被盗窃团伙顺走了。面对这种彻底扭曲毛利结构、让公司净亏损的“灾难级账本”，财务与运营部门最终妥协，同意加装锁具以求实现盈亏平衡。此外，Home Depot 约 50% 的营收来自于专业工匠与承包商（Pros）。如果一名工人在施工现场电钻突然损坏，特意赶到门店却发现货架被洗劫一空，将造成灾难性的客户流失。为了在防盗与体验之间寻求折中，Home Depot 在锁具上部署了数字化二维码，顾客扫码即可在内部系统直接呼叫店员，目前平均解锁等待时间已压缩至约 90 秒，未来甚至计划允许忠诚度计划中的高信用专业客户直接通过手机扫码自助解锁。 ### `44:59` 加锁反倒提振销量：库存有货才是硬道理 直觉往往让人以为，一旦把商品锁进柜子，繁琐的购买流程必然会导致销量暴跌。如果顾客不想在药店等待，他们大可扭头去 Amazon 线上采购；但在建筑工具与工程应急物资领域，Amazon 无法满足工地上“即时取货开工”的刚性时间窗口。实际数据显示，虽然某些品类在加锁后销量确实受到影响，但在很多高价值工具品类中，**锁进柜子后的销售额不仅没有下跌，反而出现了明显增长**。 这一反直觉现象背后的逻辑其实非常残酷：在加锁之前，那些高价值电钻由于失窃极其猖獗，货架常年处于被搬空的“假性缺货”状态，诚心买单的工匠根本买不到货；而在加上物理防盗锁后，货架上终于能够长期维持真实的物理库存。“货在架上”保障了即时履约率，反而释放了被盗窃压抑的正规销售潜力。与此同时，在全渠道与电商端，盗窃团伙则演化出了其他作案手法： - **支付欺诈**：利用盗刷信用卡、伪造或洗白礼品卡在电商下单提货。 - **退货掉包欺诈**：在线上购买正品后取出高价值核心部件，用旧机器、杂物或劣质仿品重新封装退回，利用退款审核漏洞套取全额退款。 ### `46:17` 供应链与铁路货运的“精准洗劫”：从假提单到内鬼情报 除了暴露在公众视野下的门店端盗窃，真正让大型零售商面临巨额损失的，是发生在上游分销网络与干线物流中的隐蔽截流。犯罪团伙在供应链中的渗透主要通过三种运作模式展开： ```mermaid flowchart TD accTitle: 零售供应链盗窃的三大主要通路 accDescr: 展示整车掉包、干线中途撬窃以及伪造提单提货三种作案模式的演变路径 subgraph Port["港口与转运设施"] A["进口集装箱到港"] --> B["纯粹货物截留"] B --> C["整车拖车直接掉包 / 倒运至走私货船"] end subgraph Transit["干线运输中转"] D["卡车司机长途运输"] --> E["司机按规在休息区睡眠"] E --> F["盗贼撬开后车厢快速卸货"] end subgraph Warehouse["仓储提货端"] G["犯罪团伙利用 AI 生成高仿凭证"] --> H["出示虚假提货单据与提单 (BOL)"] H --> I["冒充合法司机直接挂走整部挂车"] end ``` - **口岸与转运截留（Pure Diversion）**：在货物从供应商运往美国本土的中转仓库或海运港口时，犯罪集团直接伪造物流路径，将整车拖车或集装箱截留运往秘密窝点，甚至重新吊装上外轮走私出境。 - **干线中途撬窃**：长途卡车司机依法在行驶 8 小时后必须进入服务区停车休息，盗贼往往趁着司机熟睡的数小时内强行撬开货厢后门，迅速成箱转移高价值货物。 - **伪造提单欺诈（Fraudulent Bill of Lading）**：犯罪分子利用 AI 图像与文本生成技术，伪造以假乱真的电子或纸质提货凭证（Bill of Lading）。司机套牌开着牵引车径直驶入物流园，向仓储人员报出货厢编号并出示公章与单据，就能大摇大摆将整部拖车直接挂走。 在过去五到六年里，铁路货运编组站（Rail Yards）更是沦为有组织盗窃的新灾区。大量货运列车停放在偏僻荒凉的死轨（Dead Track）上，犯罪团伙成群结队直接切断锁具搜刮车厢。令人震惊的是，这些团伙表现出极高的“情报精准度”（Intelligence）：他们绝非盲目开箱盲盒，整列数十节车厢中，他们能准确避开低价值建材，专门撬开精确装载着枪支、精密电子设备或智能联网工具的特定车厢。零售与物流安全专家高度怀疑，航运、货运卡车甚至铁路内部已有工作人员被犯罪网络收买，为其通风报信提供精准货运清单。针对单据伪造等手段，虽然零售业正加速向电子加密提单转型，但高科技造假与内鬼渗透依然让干线防线面临严峻挑战。 ### `50:00` 利用 AI 与边缘视觉分析压缩案件调查周期 面对零售犯罪调查中海量的潜在嫌疑人与线索，家得宝（The Home Depot）并不将人工智能视为主动的“进攻性武器”，而是将其当作提升调查效率的关键工具。家得宝资产保护副总裁斯科特·格伦（Scott Glenn）坦言，摆在团队面前的案件数量远超目前的调查承载能力，"这是一场关于效率的博弈"。在传统模式下，一起典型的有组织零售犯罪（ORC）案件从最初发现可疑迹象、深度调查、立案结案到最终正式移交给警方，全流程平均需要耗费大约 70 天。通过缩短并压缩这一冗长的周期，调查团队便能在单位时间内周转并解决更多案件。 为了实现这一目标，家得宝近期对全美所有 2,300 家门店的视频管理系统（VMS）进行了彻底升级，新系统支持在摄像头前端部署边缘分析（Edge Analytics）。虽然现阶段的算法尚未普及到能仅凭一连串动作就全自动断定某人为罪犯并抓取身份信息（部分原因在于门店内覆盖高精度算力的摄像头密度仍有限），但现有的边缘视觉算力已经能够实时捕捉诸如异常逗留、刻意躲避店内员工或擅自闯入禁区等高危可疑行为。与此同时，在外部渠道监管上，家得宝也利用专门的网络爬虫技术全网监测 Facebook Marketplace 等二手交易平台，寻找疑似被盗商品的销售痕迹，通过多维度技术手段全力压缩案件的前期调查时间。 ```mermaid flowchart LR accTitle: 有组织零售犯罪调查流程与技术介入 accDescr: 展示从实体店内边缘分析监控、线上爬虫取证，到跨店调查与最终移交警方的效率优化闭环。 subgraph StoreLevel ["实体门店内线索采集"] Cam["2,300家门店边缘分析摄像机"] --> Behavior["识别滞留/绕开店员/异常移动"] end subgraph OnlineLevel ["线上二手平台监测"] Crawler["自动化网络爬虫系统"] --> Trace["追踪 Facebook Marketplace 赃物"] end subgraph CoreAP ["资产保护中心"] Behavior --> APInvest["汇总证据与交叉比对"] Trace --> APInvest APInvest --> Compress["压缩 70 天典型调查闭环周期"] end Compress --> Police["移交执法部门与检察机关起诉"] ``` - **全美门店系统翻新**：家得宝已在其全部 2,300 家门店统一更换最新版视频分析与管理系统，全面上线边缘计算能力。 - **调查时间成本**：典型的有组织零售犯罪案件从侦测到结案移交警方耗时约 70 天，技术应用的核心宗旨就是缩短这一时间跨度。 - **线上监控手段**：依靠专属的自动化爬虫抓取工具，全天候筛查主流二手转售平台上的可疑商品库存与转卖信息。 --- ### `52:15` 跨州犯罪定性与《打击有组织零售犯罪法案》（CORCA）立法转折 针对有组织零售犯罪所带来的外部性痛点——不仅严重侵蚀零售商利润、破坏正常消费者的购物体验并推高货品价格，还迫使纳税人为地方警力扩充买单——斯科特·格伦对全美范围内的法治转轨表达了强烈的乐观态度。他指出，零售盗窃早已脱离了单纯的本地治安范畴：犯罪团伙往往在佐治亚州科布县（Cobb County）作案后，迅速将大批赃物运往南卡罗来纳州或北卡罗来纳州进行销赃。依据美国法典第 18 篇第 2314 条（USC 18 § 2314）关于跨州运输被盗财物的规定，此类行为在法律上已构成典型的联邦跨州贸易犯罪，然而过去联邦层级却极度缺乏专门配置的针对性执法资源与统筹中心。 这一局面正在迎来根本性转折。继规范在线第三方卖家的《消费者知情法案》（INFORM Act）出台后，全美零售业目前正全力推动《打击有组织零售犯罪法案》（Combating Organized Retail Crime Act, 简称 CORCA）。该法案展现了罕见的跨党派共识，在众议院获得了上百至两百余位跨党派议员联署，并已作为修正案附加进《国防授权法案》（NDAA），正静待参议院最终审议；总统此前也已明确表态将签字批准生效。CORCA 法案将有组织零售犯罪、货运供应链盗窃（Cargo Theft）以及礼品卡与支付欺诈整合成一体化打击对象，确立国家级的专门协调中心并指派专属检察官，彻底改变了七年前面临非罪化思潮和“削减警费”（Defund the Police）呼声时的悲观困局。 - **法案推进现状**：CORCA 法案已通过众议院并依附于 NDAA，获得参众两院广泛两党支持，白宫已承诺签署成法。 - **州级行动网络**：全美目前已有大约 35 个州的州总检察长（涵盖民主党与共和党主政州）牵头建立了专门打击有组织零售犯罪的工作组（Task Forces）。 - **管辖权与资源整合**：法案修改了相关联邦刑事条款，正式将跨州流窜的团伙盗窃和转运定性为联邦重点打击对象，结束过去基层警力孤军作战的局面。 --- ### `54:45` 人脸识别争议与公共监控的隐私权衡 当被主持人问及目前人脸识别技术的实际发展现状，以及普通消费者走进大型零售场所时究竟应抱有怎样的隐私预期时，斯科特·格伦直言"这是一个可以再展开讨论两个小时的话题"。虽然业界基于计算机视觉的分析技术在人员识别与特征提取方面已经发展得极其敏锐，但出于公司技术保密考量，他并不便透露店内具体部署的技术产品与供应商细节。 格伦坦承自己并非人脸识别算法的技术专家，但强调社会各界围绕生物特征识别展开的探讨是完全正当且必要的。针对公共场合的无感监控，公众有权就隐私边界的划分、监控技术的潜在滥用风险，以及在打击违法犯罪与保障公民基本自由之间如何权衡利弊等议题进行审慎的公共辩论。 --- ### `56:15` 主持人复盘：技术至上主义、身份锁定与“打地鼠”的防盗困境 访谈结束后，两位主持人特雷西·阿洛威（Tracy Alloway）与乔·威森索（Joe Weisenthal）对整场对话进行了深入剖析。特雷西表示，这次访谈颠覆了她此前对“货架上锁”政策的直观感受：原先她以为给牙刷等日用品上锁会导致因体验变差而流失销量，消费者会转投亚马逊等电商；但现实逻辑在于，如果开架陈列的商品早已被不法分子搬空，货架上本就无货可卖，那么本应达成的销售机会从一开始就已归零。不过，这种防范手段演化出的科技前景也略带“反乌托邦”色彩——摄像头通过边缘算法时刻评估行人的逗留时间与危险系数；更有零售商探索通过让会员扫描 App 中的二维码来遥控解锁货架，这实质上等同于在实体零售空间强行推行某种“身份准入系统”。 ```mermaid flowchart TD accTitle: 零售防盗中的恶性循环与打地鼠效应 accDescr: 描绘防盗措施如何引发技术升级与责任风险，并最终导致犯罪形态迁移的恶性循环。 Crime["有组织零售盗窃泛滥"] --> Lock["物理上锁 / 边缘监控追踪"] Lock --> IDFriction["扫码解锁货架 / 形成实体身份准入壁垒"] Lock --> SecurityLiab["安保人员面临人身安全与诉讼免责难题"] SecurityLiab --> PoliceStrain["调查严重依赖警力与检察官积案消化"] PoliceStrain --> Morph["犯罪手法转移：转战货运劫持、线上退款欺诈"] Morph --> Crime ``` 乔对此指出，技术本身的推进往往是不可逆且自驱的，一旦能实时识别可疑行为的算法成熟，不管公众如何抱怨侵犯隐私，商家最终都会选择部署。然而技术的真实效能依然存疑：即便在每件工具上贴满防盗标签、摄像头无死角监控，但在员工与安保人员因人身安全和法律诉讼风险无法直接现场抓捕窃贼的前提下，后台证据的搜集依然不得不仰仗早已满负荷运转的地方警局予以立案。两位主持人以打趣的方式感叹这宛如一场无休止的“打地鼠”游戏：从货架盗窃转向供应链货运截击与网络欺诈；哪怕未来引入没有生命危险的安保机器人，机器人本身也可能需要购买巨额责任保险，甚至连机器人的供应链都会成为犯罪集团下手的目标，犯罪形态总能“找到自己的生存之路”。节目在对播客制作团队的致谢与常规订阅提示中正式结束。 ## Notable quotes > `04:55` 将价值观置于胸膛左侧靠近心脏的位置 — Scott Glenn > `10:00` 我们自己造成的与别人对我们施加的 — Scott Glenn > `21:37` 替警察把前期的苦活累活干完 — Scott Glenn > `24:00` 我很欣赏你的思路，但那真是一个非常昂贵的思路。 — Scott Glenn > `24:00` 虽然我们渴望将损耗清零，但最终能达到的现实数字总会高于零 — Scott Glenn > `26:40` 我宁愿每天用货损换取人员的安全，公司的执行领导层和董事会百分之百支持这一决定——我们绝不会把员工和顾客置于危险之中。 — Scott Glenn > `42:35` 为了达成 1 次真实销售，门店平均需要向供应商采购 4.2 次该商品 — Scott Glenn > `50:00` 这是一场关于效率的博弈 — Scott Glenn > `54:45` 这是一个可以再展开讨论两个小时的话题 — Scott Glenn ## Mentioned The Home Depot (家得宝), Scott Glenn, Tracy Alloway, Joe Weisenthal, Reid Lustig, Truck Parking Club, Duane Reade, Target, Kohl's, Sears, FBI, CVS, Milwaukee (美沃奇), Amazon, eBay, Facebook Marketplace, Craigslist, Meta, Asana, Cincinnati Insurance, Carol Masser, INFORM Consumers Act (消费者知情法案), Combating Organized Retail Crime Act (CORCA / 打击有组织零售犯罪法案), National Defense Authorization Act (NDAA / 国防授权法案), Retail Industry Leaders Association (RILA), National Retail Federation (NRF) ## Video description <p>If you walk into a local pharmacy in New York City, more likely than not, lots of the merchandise is locked behind closed doors as a measure to mitigate shoplifting. So how big is the &ldquo;retail shrink&rdquo; problem really? And why are stores willing to sacrifice customer experience to protect their inventory? According to today&rsquo;s guest, the Home Depot&rsquo;s VP of Asset Protection Scott Glenn, shoplifting is just the tip of the iceberg when it comes to retail theft. What is really bedeviling companies like the Home Depot is something called organized retail crime &mdash; heists by criminal outfits that seek to nab large quantities of merchandise that eventually ends up in the black market. Today, we speak with Glenn about the in-and-outs of organized retail crime, the retail shrink problem, and what stores are doing to bring down theft right now.</p><p>See <a href="https://omnystudio.com/listener">omnystudio.com/listener</a> for privacy information.</p> [Source](https://podtrac.com/pts/redirect.mp3/tracking.swap.fm/track/UVBrz8bN8aM2Xe47PEPu/traffic.omny.fm/d/clips/e73c998e-6e60-432f-8610-ae210140c5b1/8a94442e-5a74-4fa2-8b8d-ae27003a8d6b/ac7406bd-0582-4820-af93-b4c100274867/audio.mp3?in_playlist=982f5071-765c-403d-969d-ae27003a8d83)

[4:55] Scott Glenn: 将价值观置于胸膛左侧靠近心脏的位置 [10:00] Scott Glenn: 我们自己造成的与别人对我们施加的 [21:37] Scott Glenn: 替警察把前期的苦活累活干完 [24:00] Scott Glenn: 我很欣赏你的思路，但那真是一个非常昂贵的思路。 [24:00] Scott Glenn: 虽然我们渴望将损耗清零，但最终能达到的现实数字总会高于零 [26:40] Scott Glenn: 我宁愿每天用货损换取人员的安全，公司的执行领导层和董事会百分之百支持这一决定——我们绝不会把员工和顾客置于危险之中。 [42:35] Scott Glenn: 为了达成 1 次真实销售，门店平均需要向供应商采购 4.2 次该商品 [50:00] Scott Glenn: 这是一场关于效率的博弈 [54:45] Scott Glenn: 这是一个可以再展开讨论两个小时的话题

[10:00] Scott Glenn: things we do to ourself [10:00] Scott Glenn: things people do to us [23:45] Scott Glenn: really expensive thought process [26:15] Scott Glenn: I'll trade the shrink for the safety of the people every day of the week. [35:00] Scott Glenn: If Home Depot sells this combination drill kit for $499, how in the heck is Tommy1234 selling it for $200? [35:00] Scott Glenn: whack-a-mole [50:00] Scott Glenn: more cases than we can investigate

[0:00] Did you ever notice how you spend hours shopping online, only to pause at checkout because you wonder if you trusted enough to hit buy now? Agentic Commerce is testing that moment more than ever. That's where PayPal comes in. With 25 years of checkouts, 400 million consumer accounts globally, and the benefit of fraud protection. So no matter where a purchase starts, it ends with trust. Built for payments, growth in a Gentic. PayPal Open. Built for all business. Visit paypalopen.com. Some people treat ChatGPT like some kind of smart search engine, and some use it to get work done. ChatGPT Work is a new way of working in ChatGPT that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work. It's designed to help you move from a chaotic starting point to a reviewable first version. So all the source materials, briefs, and scattered information that you have to grind through to turn into something useful Something useful. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com by selecting Work Mode, available on Plus and Pro plans. At Venture Global, we think about what can be done, not what's usually done. Through innovation, Venture Global is not only building some of the largest energy facilities in the world right here in the United States, in a fraction of the time. So while others are busy talking, we're busy building. That's Venture Global. That's unstoppable energy. Hello, OddLods listeners. I'm Joe Wiesenthal. And I'm Tracy Allaway. We're the hosts of the OddLods podcast and we've got something exciting for you. That's right. So one of the best parts of hosting our podcast is we get to actually meet and interact with our listeners. And we know we have some listeners over in Los Angeles. That's right. So if you're in L.A., we're going to be recording a live show, some live recordings. And we're going to be recording a live show, some live recordings. at the Vermont Theatre in Hollywood on September 17th. We have some really exciting guests lined up, have some really great conversations planned. So go ahead and get your tickets. You can find those over at Bloomberg.com forward slash oddlots or click the link below in the show notes and come and say hi when you're there. Bloomberg Audio Studios. Podcasts. Radio. News. Hello and welcome to another episode of the All Thoughts podcast. I'm Tracy Allaway. And I'm Joe Weisenthal. Joe, we did an episode recently with Reid Lustelow, the chief marketing officer over at the Truck Parking Club. And that was a great episode about the rebound in trucking. hand way. And it really... piqued my interest. He basically said that cargo theft had become this huge deal. And it wasn't just, you know, some guys stealing boxes off the back of trucks, which is kind of what I have in my head. It was people spoofing actual carriers or pretending to be a carrier and then picking up entire cargo loads of inventory and presumably reselling it. No, that totally blew my mind. I mean, it makes sense, right? Like, OK, if we're talking in the U.S., there's tens of thousands So the degree to which any shipper could really know any carriers probably kind of loose, et cetera. It's like, yeah, why not just sort of like spring up one day and, you know, get some shipments and then take them on your own way. It makes a lot of sense, but definitely not what I would have had in mind, as you said, about what a cargo theft meant up until that point. Well, since then, I keep seeing these headlines popping up. There was a great Bloomberg story the other day actually called The Great Modern Train of railway cars. And so you can see, like, this seems to be a big deal. We know, living in New York, that it feels like shoplifting has become a bigger deal in recent years after the pandemic. And we can debate whether or not the statistics bear that out. But certainly, if you walk down the street and you go into our local Duane Reade, everything's locked up, right? Right. That's the sort of core thing is like, I don't, you know, there's probably various ways to massage statistics and all kinds of stuff, but whatever, it's not good. I see them locking the stuff up. They're locking it up. They must know that customers, by and large, hate it, right? They must know that it decreases labor productivity, having to have staffers in the store. You have to ring the bell and wait. Yeah, get a deodorant. They can't think I presume there are some real issues. [5:00] Well, we're going to get into all of it today. We'll do the sort of low level locking stuff up on shelves all the way to the high level pretending to be an entire trucking company and stealing someone's inventory that way. And I'm glad to say we have the perfect guest to try to understand all of this. We're going to be speaking with Scott Glenn. He is the vice president of asset protection over at the Home Depot and an organized retail crime expert. So truly the perfect person to speak to. Scott, thank you so much for coming on Odd Lots. Yeah, thanks, guys. Thanks for having me. I appreciate the expert designation. I'm not so sure I can live up to that, but I'll do my best. I love that you're wearing the Home Depot apron as well. Yeah, look, it's part of our culture. It's part of what we are. You can see our values over our heart is an intentional thing. Today's earnings day for Home Depot, and so we're all supporting the cause. All right. First question, and then we can get into maybe more serious topics, but what's the most stolen item from Home Depot? volume or dollars well let's do both let's do both so i would say volume it's typically something like a power drill um you see a lot of that um and then when you start to get into dollars Believe it or not, copper and wire have a lot of the commodity value out there. And so we see... People come in, they'll take a lot of that stuff out of the building and try to resell it or use it on jobs, things like that. But yeah, so certainly those are those are certainly two of the bigger ones. Joe, someone broke into there's a museum close to where I am in Connecticut called the Nathan Hale Museum, an old colonial homestead. Someone broke into it recently and stole a bunch of copper pipes. Are you serious? Like they were like old pipes that were. It happens everywhere. When I joined Home Depot eight years ago and I moved into my house between the time I closed on my house and the time that I moved into the house, they took all the copper out of the HVAC units and all the pieces of it. So it is definitely a commodity that fluctuates in price. And when the price goes up, it becomes really valuable to the bad guys. career, what your role is, why we're talking to you, like what's been the main thing that have landed you in this chair? Yeah. So I would tell you, surprised that I'm in this chair. I went to school to be an architect when I was going into college and realized that I wasn't very good at math. So I decided to maybe go work for the FBI and I got into the criminal justice side of the business. I was always fascinated by the law. I ended up going to law school after college and did that route, but started working with the Treasury Department right out of law school and worked for the Bureau wasn't my gig. It wasn't what I wanted to do forever. And so... I had during summers during school, I had worked at Target and caught shoplifters. So just kind of a side, you know, make some money gig. And luckily, I was able to get back my foot in the door with Target after I left the government and work for them and just work my way up through the ranks. So been here, been at Kohl's department stores for 10 years. I was at Sears for 10 years as the CSO. And then I came back here to Home Depot eight years ago. So before we go any further, why don't covers a range of activities from You know, someone innocently, perhaps forgetting to scan their deodorant at self-checkout to really organized, seemingly very ambitious heist. What are we talking about exactly here? How do you kind of categorize retail theft in your head? It's a spectrum, to your point. There's definitely low-level opportunistic, I'll even call it the need shoplifters that are out there. They just can't afford things. And we know that happens. And that's not... what I would say, acceptable to us, but it's also not our focus. It's not the place where we spend a lot of our energy and our tactics up to the most sophisticated criminal organizations that are out there. And I mean, sophisticated, cartel-related, transnational, gang-related, international organizations whose sole existence is to monetize theft to do other bad things, whether that's money laundering, drug trafficking, human trafficking. And these are absolutely known So there's a lot of space in between that, and it runs the gamut, but we tend to focus our time and our energy and our resources on the higher end. While we're in the business of defining terms here, I think this is helpful. A word like shrinkage is one that comes up. Would shrinkage only apply to, okay, the person who walks into a store and manages to walk out with a power drill or something like that? Or is it counted in shrinkage statistics if the... [10:00] It's like a very complex heist of an entire truck that never makes it to the destination. When we talk about shrink in retail, we're talking about, at the end of the day, I have inventory on my books. And when I physically go to count that inventory, it's not there. There's a lot of different ways that could happen. That is not just the bad guys coming in and stealing stuff. So when we talk about things within our world here, when I talk with our leadership and I talk to our board, we break it out into what we call operational or controllable administrative types of strength and then what we call the malicious. And another way to look at that is might be like things we do to ourself versus things people do to us. And that's the way we tend to categorize that ratio. And for us, I'm not going to say this is for everybody at retail, but for us, we're pretty good at controlling the operational side of the business. So about 30% of our shrink falls into that operational, administrative, things we do to ourself type of category. And then the balance of it falls into more of the malicious side or things that people do to us. on how much happens in store versus somewhere else, whether it's one of the distribution center, warehouse, et cetera, versus somewhere else on the supply chain? Yeah, so we actually break out our shrink in the supply chain separately. Then we break out our shrink in the retail environments versus the way we break out our shrink in our online environment. So we actually have three different buckets of where the shrink can happen that roll up to our larger. Now, in the supply chain, you can have operational, financial, and malicious shrink. In the stores, you can have operational, financial, and malicious shrink. And online, you can have operational, financial, and malicious shrink. So that all rolls up to the big number. But the biggest chunk of it, to your point, does exist in the 2,400 stores, retail stores that we have out there. Because again, you see these statistics, you see the headlines target closing two stores because of rising crime, or you see a company come out and say, oh, our shrink has gone up this year, but To your point, it could be because maybe they're just worse at keeping track of their inventory than they were the year before. So give us some context around how big a problem this is and whether or not it's been getting worse. You're getting to probably the most debated question in retail asset protection, loss prevention, which is... And, you know, I wish I could give you a clean answer on that. I can't. I can tell you, maybe I'll start at the top and say, How big is it? It's in the billions. It's in the tens of billions. The shrink problem. There's a lot of different estimates out there. Our cohorts, my peers, my friends and peer retailers... We all talk about this. We work with Rilo. We work with NRF. We work with all to try to get the best estimates on what this number is, but how people categorize shrink. What somebody calls operational strength versus administrative versus malicious can vary by retailers. And so you've got the retail component. We are a retail accounting operation. There are companies that are on cost accounting. And so making that transition between the dollars and how do you include markup and to get to a number, very, very complicated number. What I can tell you is that... There's these numbers that have been around as long as I've been doing this in retail, which is like 40% of shrink on the malicious side is done by associates working for the company. They cause 40% of your loss. 35% are people walking into your store and causing it. And 20, 25, whatever the balance is, is that self-inflicted operational administrative types of shrink. And so... Those numbers add flow. They go every year. There's surveys that are done. They might go up a couple of percentage. They might go down a couple of percentages. what I can speak to, is... When I got here at Home Depot in 2018, I wasn't even really a big believer in what you would call the organized retail crime problem. also pre-COVID, also before Omni really took off and has blown up to what it is here. Places where you can resell very impersonally. Wait, what's Omni? Online. Online sales. Sorry. Omni channel. Sorry. But when you think about... the kinetics of all that and how that's changed over time. And I would tell you that I have become a believer in one organized retail crime exists. It's become more lucrative. It's become more pervasive. And it's become more dangerous when it comes to how it impacts the customers and the associates in the stores. So, While those numbers have generally estimated state in the kind of same buckets, I believe they're starting to drift up. And I also believe since COVID, they've been underreported. And so I think that there's a big piece of that, that it's just not necessarily reflecting in all the numbers that you see showing up in police reports. [15:00] FBI uniform crime reports, things of that nature. Did you ever notice how you spend hours shopping online, only to pause right before you hit buy now? That tiny hesitation, the one where you wonder if it is trustworthy, can make or break the sale. Now, with AI changing the way consumers discover and compare things, that split-second trust question matters more than ever. That's where PayPal comes in. For more than 25 years, PayPal has been a leader in online payments. And now they're at the forefront of agentic commerce, making it work for businesses, letting them maintain control of their brand and their customer relationships. most still feels familiar and deeply dependable. Build for payment, growth, and agentic. PayPal open. Built for all business. Visit paypalopen.com to get started. This is the Bloomberg Tech Minute brought to you by ChatGPT. Now with ChatGPT work. I'm Carol Masser. Has the AI moment arrived at corporate offices worldwide? Not yet, and it's perhaps several years away, according to one survey of corporate recruiters who circle the world's business schools each fall and spring. including MBAs, were communication, problem-solving, and adaptability, which changed little from last year. Training in artificial intelligence tools ranked in the bottom third of the skills list, but recruiters did say AI-specific training will be more important in five years. On that, recruiters rated current graduates as only somewhat prepared or less to work with AI tools, and said business schools are not doing a good job of teaching AI skills. Bottom line, according to The only thing you want when it comes to AI skills is for candidates to be able to automate routine work. That's the Bloomberg Tech Minute brought to you by ChatGPT. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com today by selecting Work Mode, available on Plus and Pro Plans. If you listen to financial news, you know a lot of time is spent thinking about what's next, the next opportunity, the next investment, the next move. But sometimes what matters most is being ready for what you never saw coming. A storm damages a second home. A pipe bursts above a collection you spent decades building. An unexpected loss puts a growing business on pause. Those are the days no one sees coming. But for more than 75 years, Cincinnati insurance has been ready when they do. With deep expertise, coverage for businesses, homes, valuables, and more, Cincinnati Insurance works with independent agents who take the time to understand what matters to you. Together, they help protect more than what you own. They help protect what you're building toward. And when a bad day does come, Cincinnati Insurance has real people there to help make things right. Because planning for the future isn't only about knowing what's next. It's about making sure you're ready for what you can't predict. Let Cincinnati Insurance make your bad day better. Find an independent agent at CINFIN.com. So maybe this is a good opportunity to sort of dive into some details. But you said earlier in the episode that you're more interested in the sort of more sophisticated, organized drivers of shrink than perhaps a random person who maybe they can't afford an item, etc. Can you talk to us about In a given theft, et cetera, someone goes into a store and they walk out or they take something that they didn't pay for. What are the fingerprints that look like, OK, this we have reason to believe was part of something sophisticated and organized versus this is like, yeah, we don't. support that but it's one random person and it's not going to be the main thing we prioritize. Yeah, you guys asked some great questions to get right into the details of how some of this becomes difficult. So in an individual instance, it's almost impossible. You know, if you talk to somebody... If the, you know, we have people in our stores, their whole job is to stop theft. That's what they do, that they detain people, they prosecute, they work with the law enforcement and they deal with that. And then we have groups of people here in Atlanta whose job is to handle the more sophisticated components of this piece of it. But it all starts locally. It all starts in the stores. And so some of that is, hey, I see somebody steal that drill and you sit down and have a conversation with that person before they end up going to jail. and they were really using it to work on a project on their deck on their house. Whereas you might [20:00] have that conversation with them and say, yeah, I was asked to come in and steal this and I was going to go take it to another place. And I was going to get money for that, or I was going to get drugs for that, or I was going to get something. And I was going to get given a list to go to another store and go get something else. And so we cultivate that intelligence, we cultivate those resources, and we report that back to our folks here who are organized retail crime experts and do this for a living, most of whom are former federal law enforcement agents or law enforcement agents at state and local agencies. And their job is to figure out which is the wheat and which is the chaff and what are the things that are the most important thing for us to go. in store B. We've got a report from that building that says, you know, Johnny with driver's license, one, two, three, four, five, six, has been in now four stores apprehended for theft because, you They're going to get apprehended. They might be put in jail and they'll be out the next day. You know, they're going to make bail or there's going to be no cash bail or whatever the case may be. And they're going to go back to doing what they're doing. And so that's how we try to, like, stitch these cases together and understand which ones are the ones that are more valuable to pull on versus the ones that are like, this is really a one time occurrence. And this is not something that we need to spend a lot of resources on. if we can identify patterns and stitch the sort of ring together. That sounds like something the police should be doing. Sometimes it's hard enough to get the police to show up. And I'm not blaming in this case. I mean, we know that police departments are under-resourced, stretched very thin. When you talk about retail theft versus, you know, robbery versus carjacking versus murder and the escalation hierarchy, right? Where are they going to spend their time, right? So they're not going to always be able to have the capability to come to our store and deal with this. So I would tell you our, I'll call it our internal behind the scenes, our job is to do the work for them. Our job is to hand them a case file that is 80% complete that says, I've given you everything that I can possibly do for you to then walk into a judge, get a search warrant, get an arrest warrant, do everything that I can't do. And they're going to respect that. I don't have to do all the behind the scenes labor. And all I've got to do is walk in front of a judge and say, this is a great case. And we believe we believe everything that's here and put their hand up and swear on the on the on the Bible. Then they're going to take those cases much more regularly than somebody that's kind of saying, hey, you just go figure it out for me. Just one more question on this. Is it true that some retailers will not stop shoplifting until they observe the shoplifter getting to a certain threshold of goods stolen? meaningful charge? You know, I would be hard pressed to speak to that for what the policies of some of my my compatriots are there. But what I would say is, is we don't do that. When people say that, I feel like they're kind of looking at things in a very narrow scope. And going back to our point earlier... This may be something where we're already working on this person. We know they've been in 50 stores over the last 30 days stealing from us. And if they walk out the door, it may be because we're going to go watch them go to where the fence is or where they're going to resell the product or where they're storing the product. That might be part of an operation versus a, hey, we're letting you go get to a threshold. And that's really what, at least from our end, we're not trying to build up big charges on people. We're trying to say, I want to get the complete picture of what's happening because when I wrap this up, I want it to be completely wrapped up. Intuitively, it seems like it would probably be pretty hard to prevent someone from stealing a washer, like the little disc or something like that, like put it in your jeans pockets or whatever. should make it very hard intuitively to steal a power drill, because I imagine you could put tags on it that would set off alarms, that you could have video surveillance that could simply set off something. If a person was in one aisle, trigger, I have to imagine it's not that hard to train a model to identify someone, take it off the shelf, and then that person makes a route out the store without ever having gone to the register. And, This seems solvable by technology alone. Why isn't it? What is it makes such that, A, it must still happen all the time, or B, it only gets stopped if a person at the end happens to notice it? I love your thought process. That is a really expensive thought process. Okay. And you think about a Home Depot store, 140,000 square feet of a working warehouse to put the cameras that I would need to do that. And by the way, there are places where we do that. There are places, you know, in some aisles, we're able to do that. But we have a lot of valuable stuff at our stores, in different parts of the store. At the front end, we have AI, you know, Basecamp. [25:00] We have a lot of different, I would say, video analytics that are continuing to evolve and we're really leaning into. But it is expensive and it's a lot of processing power. It's a lot of hardware that you have to put into your stores. We have some advantages with our scale and our size to be able to do some of that. Not everybody does. And certainly not everybody has. Sometimes it's a little bit of a tradeoff. Like how much are you willing to take and shrink versus how much you would have to invest in hardware, capital expense, operating expenses to be able to put it to zero. I mean, the number we aspire to is zero. The number we'll get to is somewhere north of that. alarm. Why at a level of a power drill is that an insufficient technology to get the number to zero? It's not insufficient. We've played around in the bricking of product space. When you think about, you know, the margins in some of these places, does it make sense to do it, to go have to redefine your entire production line for a Milwaukee drill? in order to insert a technology, a Bluetooth chip of some sort. We've tried that. It's very, very hard to scale up. And again, you still have to put hardware in your stores to be able to detect those tags. That's the piece that typically is very expensive. Again, none of this is impossible. It's just like there's always trade-offs and you're working towards that. it's going to take some time. I mean, to be honest with you, RFID has a lot of promise in terms of tracking not just the theft side of product, but like Where is this product in the store? Has it been sold? What's the replenishment look like? How do you get your on-hands corrected? There's a ton of business opportunities on those pieces, but I think that RFID hardware costs have always been so expensive, it's almost impossible to put it in a Home Depot store where we have eight separate entrances in and out of our building. How much of this is a liability issue in the sense that, okay, I can tag an item and I can hear an alarm ring if someone is going out with it without paying, but... There's still an issue over whether or not you want your employees, even your security guards, going after people who might actually fight back and be dangerous. And I'm sure we've all seen you, especially Scott, the videos of some guy who just fills a massive shopping cart and walks out of the store with the security staff looking on. It's the bigger issue than the financial, right? Like at the end of the day, there's a certain level of we understand there's going to be some shrink in our stores. There's a certain cost of doing business. When you open up to the public, you're going to have some of this, this type of pieces of it. But this is the problem specifically. with the organized retail crime folks the professional criminals A lot of times they're using homeless folks, addicted folks, people that are maybe on the down low, and they are monetizing them to go in and do this. And they're only... Thought process is, I'm leaving with this product because I need to get my money. I need to get my whatever. And they will do violent things. And they have done violent things. And not only has it happened in retail, it's happened specifically to us. And we've actually had people injured. We've had people killed. in our stores in these types of incidents. where that is just we're not going to allow that. So our security folks are asset protection people in stores. They're trained. They're trained not only on how to do what they do, but also how to de-escalate, how to observe and pull away from those situations when they start there. But we also have 400,000 store associates who are very passionate. They don't like to see stuff stolen. And we train them very clearly on like, you don't get involved with these things. You back away. And we do give them the same, you know, a slightly different, but still de-escalation type training, because that's the problem. I'll trade the shrink for the safety of the people every day of the week. And my leadership team and our board would back us on those decisions all day long. We're never, ever going to put people in jeopardy. Our associates and our customers are the number one tags for 25 years now, I would have assumed that the price of this hardware would have come down dramatically such that you slap a little tag on and it's not very easy to remove. And then by the time you go through some gate or entrance, it sets off an alarm. Can you talk about why in 2026 that technology is still actually expensive and hard to scale? No, I can't. I don't understand it. It's your lips to God's ears, right? Like it's one of those things where I'm with you. And I will say a Home Depot environment, a little different. We are a hardware store with basically a Faraday cage of steel throughout our stores. We do... Passive RFID does not work in our stores. You have to go to an active pinging RFID technology. We've tested it. We know it can work. Again, it's all about typically... [30:00] You just explained that it does not work in your stores. Actually, I don't know anything about this. Can you clarify that? It's difficult. The signals are blocked by the steel. That's typically the way it works in our stores. For a passive technology like that, there is an active version of RFID that's out there. But again, it's typically – and it's not like they're not coming down. I mean, tag costs are nil at this point. That's not really the issue. It is truly the hardware capabilities to be able to put it everywhere you want, to be able to read the way you want it to. It's still a very expensive technology to this point. Three years from now, Who knows? So why don't we talk about what happens after the goods get out of the retail environment? And maybe we should talk about the distribution environment at some point, too. But what is fencing? Not the sport. So, again, I kind of mentioned what we call the boosters. And we're specifically now talking about organized retail crime. Boosters are the people that do the stealing. Typically, I'm not going to be able to do that. They are recruited. They are given a list. They are given money. They are given, in many cases, drugs or the promise of drugs to when they come back with the product, which is kind of the hook to make sure that they don't just walk away with the product and their money. And these folks are the ones that go out and they take the risk. They pull the product out of the stores and they do it. Like I said, they can be very aggressive. They could be very violent in many cases. And all they're thinking about is how do I get out of this building and go to where I'm supposed to go? And they take it to what we call the fence. The fence is kind of the consolidator of the product. So this could be somebody's house in their garage. This could be a warehouse facility. This could be a flea market. This could be you name it. I'll get back to there's another interesting twist on this piece of it that I'll get back to what the boosters are doing and fences are doing here. But then these fences are the folks that actually clean the product and then offer the product for resale. So when I say clean it, to your point, some of these have electronic tags on them. They might have identifying this is a product that's sold at Home Depot. This is a product that's sold at Target. This is a product that's sold at CVS. Stickers, identifying price tags, things like that. They want to clean all that product. and then they'll offer it for sale somewhere. More and more since late teens, early 20s, that's online. In the past, I would say it was some of the bigger online folks, the Amazons of the world, the Ebays of the world, the bigger places. Since the Informed Consumers Act was passed in 2023, I believe is when it came effective, we've seen less of that on the bigger places. And now it's moved to more of the peer-to-peer markets. So the Craigslist, the Facebook marketplaces of the world. that's where a lot of that product is then offered for sale. And so that's the fencing operation. And then the money is then kicked up to wherever or at least a portion of it, just like your traditional organized retail or organized crime. It's kicked up to some bosses that are managing this whole operation from the top. And whether that's a five guys that are sitting in an apartment or a house that just kind of decided that they want to be organized retail criminals. Certainly fits the definition of all those types of things. And they're using the money to furnish a lifestyle or a cartel or a Chinese transnational gang or a South American whatever. That could be any of those people that are the ones that are sitting at that top level. supervising the fences, making sure that the product's getting moved and making sure that the money's coming in. So I've heard, you know, there have been stories about Facebook Marketplace and frustration that Facebook hasn't done more. I'm not sure what the status of that. You know, if you see for sale on Facebook Marketplace, like a 10 pack of unopened Hanes t-shirts or something like that, there might be legitimate reason to think that this was stolen. It could also be a for them like it's plausible. What is the optimal role for an owner of a peer-to-peer platform to play? And then what do you perceive right now is the gap between what would be a sort of like optimal role for them and contribution to fighting this versus what they're doing right now? The short answer is it's very expensive to patrol your platform, to know your seller, to make sure that you really understand that they're a legitimate business and that they're a legitimate reseller. inform the federal bill that was passed earlier required the traditional marketplaces to actually Do that. Take that step and make sure that they did. Just like I have to know who my suppliers are here at Home Depot. They had to know who their suppliers are, the Amazons and the Ebays of the world. There was a loophole in that bill that did not cover the peer-to-peer resellers. Can't explain why it happened. Just it happened. And so that loophole has continued to exist for it. And they're not going to put the governance in to do it unless they're forced to, just like the bigger boxes didn't have to before in form. And so we spend a lot of time governing those marketplaces. So we have... [35:00] of technologies that are looking at that and say, If Home Depot sells this combination drill kit for $499, how in the heck is Tommy1234 selling it for $200? And at volume, how can he sell new in-box product at a price lower than my cost? And so we will look for that type of stuff. And that's where a lot of our investigations also will start. We'll actually start on the marketplace and work backwards to the person coming into the store, going back to kind of our earlier discussion. But it's whack-a-mole to a certain degree, right? I mean, we're really patrolling millions and millions of sellers that are out there. So we build criterias. We build models that kind of look and say... boy, this person sells a lot of drills. This person sells a lot of drills new in box. This person sells a lot of drills new in box under our cost for the product. So that's kind of the way we tend to do that. But it's an uphill climb. It's a big battle. This is the Bloomberg Tech Minute brought to you by ChatGPT. Now with ChatGPT work. I'm Carol Masser. Has the AI moment arrived at corporate offices worldwide? Not yet, and it's perhaps several years away, according to one survey of corporate recruiters who circle the world's business schools each fall and spring. including MBAs, were communication, problem-solving, and adaptability, which changed little from last year. Training in artificial intelligence tools ranked in the bottom third of the skills list, but recruiters did say AI-specific training will be more important in five years. On that, recruiters rated current graduates as only somewhat prepared or less to work with AI tools, and said business schools are not doing a good job of teaching AI skills. Bottom line, according to The only thing you want when it comes to AI skills is for candidates to be able to automate routine work. That's the Bloomberg Tech Minute brought to you by ChatGPT. Put ChatGPT to work on your most ambitious ideas and projects. Get started at ChatGPT.com today by selecting Work Mode, available on Plus and Pro Plans. If you listen to financial news, you know a lot of time is spent thinking about what's next, the next opportunity, the next investment, the next move. But sometimes what matters most is being ready for what you never saw coming. A storm damages a second home. A pipe bursts above a collection you spent decades building. An unexpected loss puts a growing business on pause. Those are the days no one sees coming. But for more than 75 years, Cincinnati insurance has been ready when they do. With deep expertise, coverage for businesses, homes, valuables, and more, Cincinnati Insurance works with independent agents who take the time to understand what matters to you. Together, they help protect more than what you own. They help protect what you're building toward. And when a bad day does come, Cincinnati Insurance has real people there to help make things right. Because planning for the future isn't only about knowing what's next. It's about making sure you're ready for what you can't predict. Let Cincinnati Insurance make your bad day better. Find an independent agent at CINFIN.com. That's the gap Asana is built to close. Asana is the operating system for human agent teams, your easy button for AI productivity across every team. Ready-to-go AI teammates pre-built for marketing, ops, and IT. No prompt engineering, no setup. They show up where the work is happening, already onboarded in your workflows, ready to deliver. And the more they work with your team, the smarter they get. With Asana, your whole company can work on the same plan towards the same goal, or 10,000. Asana, where humans and agents workflow together. Try it at asana.com. That's A-S-A-N-A dot com. This might be a very niche question, but do you observe any interesting trends when it comes to the prices that fences are actually choosing for these items? Because they could I mean, if you have something new in a box, you could maybe try to charge almost full price for a power drill or something. Or you see them with like massive discounts as well, and they're leaving money on the table. I wouldn't tell you that I've done a lot of research into what the specifics are. [40:00] consensus is The more technology-laden the product is... the more it sells closer to retail. That plus gift cards. Gift cards tend to sell very close to retail. So we see a lot of our products selling in the 70% to 80% of retail range. Whereas in some cases, you'd think about the deodorants and the t-shirts and the Tylenol and some of that tends to be in the 30% to 40 cents of retail. market. So there's some broad strokes on that, but honestly, I don't spend a ton of time doing that. It tends to be like if it's internet connected, it sells very, very close to retail. If it's technology driven, it's got some resale capability to it. Efficient markets, finding the right price of anything on any platform. So you talk about this loophole in the law that allowed the peer-to-peer sellers to What would be that additional step? OK, if they were to close this loophole, what would they then be required to do? And are you either surprised or disappointed given, I mean, a company like Meta is gigantic. Are you either would you have been surprised or a little disappointed that they would not feel the impulse to do this voluntarily without being required by law? They'll tell you that they do a lot of policing of their site. And I'm not arguing... that they don't have... I know they have departments that do that. We work with them. I mean, so they're not completely willfully blind. They're not completely out there just saying, no, it doesn't happen on our site. we think they could do more. And again, it comes back to, If they had information, if they had... policing in place that said, How in the world does this guy undercut Home Depot or Target on pricing? That should be a flag for them to say, I'm going to pull you off the site. I'm going to suspend you from the site versus us having to come to them and say, I'm going This guy is undercutting us on pricing. He's selling at volume. He's selling at scale. We've bought the product for it. We know what store it came from. We know that this guy's involved in crime. Now you go pull them off the site. And so your answer is both disappointed and surprised. Like, do I think they could be doing more? For sure. Do I think that I should be the one that has to come? It seems a little backwards to me that I should have to come and tell them that they've got somebody selling stolen product on their site versus them knowing who their suppliers are, knowing that tax records exist and that there's filings that are out there that they should be held accountable to the same way the Amazons and the Ebays of the world are. trying to shrink the shrink in various ways and customer experience because again we've all experienced the frustration of going into a usually it's a drugstore of some kind and hitting the plastic wall and having to press the button and wait for someone to come and then you're standing there for 10 minutes and then they don't have the right key and sorry this happened to me recently no keep going on it's so seriously they don't have the right key yeah stores i get a few our leadership. So I gotcha. But stores, some stores have clearly made the decision that it's worth it to them economically to do this. We have. It is not an easy conversation with your merchants. It's not an easy conversation with your operators. It's a very arduous, data-driven discussion. And so I can think back when I got here in 2018, middle of 2018, and we were on a trajectory of shrink that was not good. It was going wrong. Way the wrong way at that point. And we started to have some of these conversations. But I'm a data guy and I'm a test and learn guy. I like to have my facts. And so we did some testing. We did this. We did a lot of looking into what products we were losing very, very specifically under the microscope. we found that one particular combination of kit that we were selling for $799, we would buy it 4.2 times for every time we sell it. Now, go have that conversation with your finance folks. And they go, that's a gross margin mix we don't like. That's a real business opportunity. If you can get me to break even, if you can get me to slightly positive on that, you know, from a gross margin perspective, then they go, okay. Okay, there's a little bit of a trade-off, but it's also we're in a unique business. We're in a business that 50-ish percent of our revenue comes from pros and contractors. If they leave a job site because their drill broke and they go to the store because it said we had one and we don't, that's a problem too. And so, you know, as we kind of walk through and we work through this, we got to a very curated list of products where we said... we're going to probably have to do something. And if we do it, We can do it the right way and try to minimize the friction. We use QR codes on all of our lockups that summon somebody through our phone that they come and it takes on average about 90 seconds for that lockup to be unlocked by a person. I want to be in a place where customers can self-service. They scan the QR code and I know that they're a pro that's in our loyalty program. Why can't [45:00] to give them access to that. So we're working through the friction. That's always going to be an opportunity for us, and we're trying to minimize it to the best we can. But It's not perfect. And it's definitely a little bit of a balancing act that you walk on some of those particular products. You can't do everything. That's just buy online. If that exists, that's Amazon. If you don't want to go into a store and you don't want to be held up and you can wait 3 hours to get a product, go to Amazon. That's something we do. Now, thankfully, they don't necessarily operate in our space. Like, you know, don't sell power drills like we do. Do you notice that sales actually go down for items that have been locked up? Not necessarily. There are some categories where that is true. There's also many, many categories where actually sales go up because you're in stock. That's interesting. I hadn't thought about that, but that makes a ton of sense. It's funny when I go into pharmacies and it's all locked up, I always say to myself, I'm just going to start shopping online for all this stuff. But I never actually do. I'm always back in the pharmacy like every week. of shrink when we're talking about the e-commerce context? - There's a couple of ways. I mean, a lot of this falls back to payments fraud, credit card fraud, gift card fraud, the payment process, oftentimes is the way that they're defrauding the omni space. There's a, to use your word, a niche piece of people buying product. taking the product out, using it, putting something back in and then doing returns. So they get they get back on the return side. So they're getting credits for the product that they actually have and then send back to get their full refund price on. So that's a part of the business that we deal with in many cases that we're working through. But it's typically around the payment side and the gift card and credit card fraud. Isn't there some like e-commerce like incidents for people where they they're like, I want to return this and they'll get the refund and like don't bother sending it back. Sure. Yeah, sure. I realize we've been very focused on the retail side of theft. But can you talk a little bit more about what happens on the back end with distribution? Because, of course, you know, you're shipping a bunch of stuff from its production center over to the store itself. I imagine there are many, many avenues along the way that a smart criminal could exploit. Yeah, it's growing, to be honest with you. I think there's, I would say three main modes. go back to the modal piece of it, but like there is pure diversion. So whether that's in route from vendor into, if you're an importer that, you know, bring it into the U S you go into your transload facilities, your port facilities, you know, There's people that divert full trailer loads back someplace else from that port, could be back onto a container shift to go right back over someplace else that's been stolen. There are break-ins in terms of, you know, actually just as truck drivers drive eight hours across country and they got to park their trailer somewhere. And during the time they're sleeping, they're breaking into the back of the trailer and, you know, just taking the boxes off the back. And then there's the one you mentioned a little bit earlier, which is the diversion on the supply chain side back out. fraudulent bill of lading, fraudulent pickup paperwork. AI is playing a space in this where they're able to duplicate some of the capacities that allow a driver to just drive straight in and say, I'm here to pick up trailer 1, 2, 3, 4, 5. And they dribble right out because their paperwork looks really real. Now, we're moving away from that with the electronic billing and some of the new ways that we're dealing with it. But it's not necessarily the scheme that's Thank you. go after. Rail has become an issue. Rail has definitely become a bigger issue over the last probably five or six years where we're seeing, you know, as the cars are parked in yards, they're being broken into by mobs of people that just kind of strip them down for what's in it. But they have really good intelligence, which is interesting. So, you know, we think they have people working inside the companies, the shipping companies, the... the trucking companies because... They don't hit every trailer or every car on the rail. They hit the car on the rail that has the product that they really want to go after. And so whether that's firearms, whether that's electronics, connected devices, you name it, there's a lot of, I would say, kind of a new avenue opening up on the rail space. And a lot of dead track, by the way, out there, too. So it's wide open. It's interesting you mentioned, you know, showing up with a fake invoice or a fake bill of lading. I don't know if you put on an orange smock, et cetera. Like how much, you know, you just like just blend right in and like fit it, fit in. And no one. No, I'm sure people are trained to be suspicious. But you mentioned AI, like how much of like the tools I'm sure there's, you know, we talked about cameras that could theoretically detect when someone has put something in a pocket. et cetera. But the tools that we call AI, have you found uses for them in [50:00] I guess sort of, offensive deployment, whether the models can autonomously just scan through Facebook marketplaces and look for anything that we Home Depot had found suspicious. How much sort of have you found that the models can be sort of applied towards general tasks and find helpful info for you? Yeah, look, I'll say it this way. We have more cases than we can investigate. And so for us, it's an efficiency play, right? So the more we can shorten the duration of these cases, like our typical ORC case takes about 70 days from discovery to closure to handing off to the police department. And so if we can compress that timeframe, we can turn more. And so we believe that the video, we just replaced our entire video analytics, our video management system in all 2,300 stores. to an upgraded version that allows us to now put edge analytics on our cameras. The ones we have, you know, doesn't get necessarily to the point where you're talking about where this action or this set of actions now becomes suspicious and identify somebody. We'll get to that. There's there's a place for that. I just don't have enough cameras in all my stores to do that. But the cameras that I do have, we are putting edge analytics around there that help us identify persons that were walking in this store and like they were lingering in this space too long or they were avoiding associates or they were in a place where they shouldn't have been. That's definitely out there. You mentioned the web crawling, you know, that's out there. That capability is out there. We have our own company that does web crawling and identifies these things that absolutely looks for all that. So anything we can do to kind of like, compress that investigative cycle. is how we're using AI to deploy against the bad actors that are out there today. Obviously, shoplifting theft, it affects all of us. It hurts your profits. It hurts my experience. It probably raises my prices, etc. There is a cost that we have to pay for taxes to increase the amount of police capacity, etc. This is a classic example of very bad externalities for almost everyone involved. forth. There's been the last month, everyone's talked about the flock cameras, etc. What do you perceive as the sort of appetite among law enforcement or the politicians that would have to set laws to actually move the dial, etc. on some of this in today in 2026? Well, what I would tell you is I am super bullish on where this is going. We talked about Inform, which was probably the first major ORC-driven act out there. We are now working on Coworka. Coworka is the Combating Organized Retail Crime Act. Home Depot, retail in general, has been a supporter. This is actually a very bipartisan act. It's passed the House. It has been attached to the NDAA. It's waiting on the Senate to sign it or to pass it. The president will sign it. He's already indicated that he will. This is literally the first organized retail crime act at the federal level that provides resources, prosecutors, coordination centers, defines changes to USC 182314 that says, here's how you have dedicated to go after this because here's the issue. This is not a local crime. This is not a – it happens in the county of Cobb here in Georgia. It goes to Cobb and it goes outside the state of Georgia and it goes into South Carolina and it goes into North Carolina. That is now a federal crime. This is interstate commerce. This is interstate transportation of stolen goods. All these things, yet there's no federal resources dedicated to this. So this gets ORC, this gets cargo theft, this gets gift card and payments fraud, all wrapped into one bill that is going to really change the way we as retailers can address this problem of organized crime. So... The short answer is absolutely, I think the tide is changing. It's not just one-sided. There's, I want to say, 100 to maybe 200-plus House sponsors, both from the Democrat and Republican side. There are many, many bipartisan sponsors in the Senate for a coworker. There are 35-ish states that have organized retail crime task forces led by their attorneys general. both in Democrat and Republican states. So, Seven years ago, I was doom and gloom on this. Nobody was taking this seriously. Nobody thought this was a big problem. This was petty shoplifting. This was not a thing that anybody really wanted to go on. In fact, in many cases, I think you guys know it was decriminalization was happening, lots of defunding the police talk that was out there. I think that this has helped us move much more in the right direction over the last couple of years, particularly. What is the state of the art with facial recognition? What are the expectations that people should know when they're like walking into a major public space, the degree to which either the cameras in [55:00] Major retail establishments either are able to or are able to identify who they are and how long that data is stored and so forth. that's a conversation that could probably go for another two hours. Yeah, I know. I would say that at the end of the day, video analytics in general are getting really, really good about recognizing people. And we don't talk about specific technologies that are out there. I'm not an expert in facial recognition, so I don't know. What I would say is that... I think it's fair to have conversations about privacy. It's fair to have conversations about misuse. It's fair to have conversations about balancing the good versus the bad that can come from any technology that's out there. But I'm probably not the person to really talk about facial recognition. All right, Scott Glenn, thank you so much for coming on All Thoughts. Really appreciate it. No, thanks for the time. I appreciate you guys. So, Joe, that was a fascinating conversation. You definitely helped me understand the whole lockup issue. that I kind of assumed that sales would go down because so many people would be inconvenienced and you just walk away and say, whatever, I'm going to get my toothbrush from Amazon. But if the inventory isn't there anyway, because people are stealing it, then that's a lost opportunity for sale. The other thing that stood out for me is there are some slightly dystopian aspects in there. So the idea of, you know, videos with edge capabilities, like might be looking at person as a risk. Like, oh, this person has been standing over in that corner for a while, looks kind of suspicious. That doesn't sound great. The other thing that was very briefly mentioned is this idea that... Well, maybe there's a way around the lockup program where you could just scan a QR code from your... I don't know, Walmart or Home Depot loyalty app, and it would automatically open the shelf for you. Which sounds very convenient, not very convenient, but more convenient if you're a shopper, a regular shopper at those stores. But on the other hand, you're kind of getting close to an almost identity system for shopping. All great points. But yes, you know, it does feel like we're probably going to increasingly be headed in that direction. Whether we want to or not, the technology itself will be the driver. itself that could just say identify suspicious behavior in real time, it will be implemented. Right. Like there is it. People might complain, but if the technology exists, it will eventually be implemented. I'm very anti-shoplifting, so I think like, you know, and like I don't really – I think, you know, things would be better if there were less of it, etc. And I'm not particularly worried about like, well, if I'm in a public space. But what if you just want to stand around for a couple hours admiring the power tools? The person comes by. No, look, there are obviously clear issues that are going to arise and they're going to be like. very uncomfortable and unsettling. But I believe that like, the technology will lead the way rather than anything that we like, prioritize. Here's the thing. I think the question you have to ask of the technology is whether it leads to good outcomes in the forms of more people being caught. Because part of this is like, okay, you can install surveillance in the store, you can stick tags on every individual item. But if security can't actually stop people from stolen goods because of liability issues, then... it's not really doing that much, right? Maybe it'll help in the investigation, but then you're still dependent on a potentially strained police force actually taking up the case and using the evidence that you've gathered. It does feel, you know, he used the term whack-a-mole, and I feel like... One of the meta phenomenons that has happened is over time shoplifting or theft has switched on these various online frauds, etc. But you would think like, oh, well, if you just get rid of the retail stores, then that takes care of the problem. But the crime finds a way and new forms of crime will emerge. And one day we'll probably have robot security guards who don't have like a personal safety issue because it's a robot and they'll like reach out and comprehend the person. But then there will be a theft on the robot supply chain and so forth. And like, it'll probably never go away. But it was interesting. Maybe the robots will need liability insurance. You don't know. They very well. They very well might. All right. The robots will be shoplift. That's right. The misaligned, misaligned robot security guard. That bolt over there is really shiny. Yeah, that's right. All right. Shall we leave it there? Let's leave it there. This has been another episode of the All Thoughts podcast. I'm Tracy Allaway. You can follow me at Tracy Allaway. And I'm Joe Weisenthal. You can follow me at The Stalwart. at CarmenEarmanDashelBennett [60:00] Cashbot Kale Brooks at Kale Brooks and Kevin Lozano at Kevin Lloyd Lozano. And for more OddLots content, go to Bloomberg.com slash OddLots. We have a daily newsletter on all of our episodes. And you can chat about all of these topics 24-7 in our Discord, discord.gg slash OddLots. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes absolutely ad-free. All you need to do is find the Bloomberg channel on Apple Podcasts and follow the instructions there. Thanks for listening. Music Healthcare doesn't always work great. If you've ever waited on a refill or couldn't schedule an appointment, you get it. That's the kind of stuff Optum is changing. They're using data and technology to integrate patient care, pharmacy, and everything else. So healthcare is connected, not complicated. What's that look like? Cheaper prescriptions that are easier to get and care that looks at the whole person how you need it. Optum is helping make healthcare work as one for everyone. Learn more at business.optum.com. One last thing before you go, you know that feeling when feedback, notes, and data are coming in from every direction? ChatGPT work can make sense of it all, turning scattered inputs into ready-to-review work. With your permission, work in ChatGPT can gather context from apps and files, summarize incoming messages, rebuild decks, and help finalize high-quality documents for your team or clients. Stay in control as you move from a goal to polished work faster with ChatGPT work. Download the ChatGPT desktop app today and take on your most ambitious work. Game Night Rush, or any night of the week really, Genius keeps every order moving. From online ordering to your kitchen to the front counter. Big League reliability for any business. That's genius.