《世界經濟論壇（WEF）：2025人工智能變革全球貿易研究報告：提升貿易技術的效率、可持續性和包容性（英文版）（36頁）.pdf》由會員分享，可在線閱讀，更多相關《世界經濟論壇（WEF）：2025人工智能變革全球貿易研究報告：提升貿易技術的效率、可持續性和包容性（英文版）（36頁）.pdf（36頁珍藏版）》請在三個皮匠報告上搜索。

1、Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTechI N S I G H T R E P O R TJ A N U A R Y 2 0 2 5In collaboration with the Ministry of Economy of the United Arab Emirates and Abu Dhabi Department of Economic DevelopmentImages:Getty Images,MidjourneyDisclaimer This docu

2、ment is published by the World Economic Forum as a contribution to a project,insight area or interaction.The findings,interpretations and conclusions expressed herein are a result of a collaborative process facilitated and endorsed by the World Economic Forum but whose results do not necessarily rep

3、resent the views of the World Economic Forum,nor the entirety of its Members,Partners or other stakeholders.2025 World Economic Forum.All rights reserved.No part of this publication may be reproduced or transmitted in any form or by any means,including photocopying and recording,or by any informatio

4、n storage and retrieval system.ContentsForeword 3Executive summary 4Introduction:AIs potential as atrade technology 5AI adoption 6AI goals in trade 71 AI:A composite technology for trade 92 Use of AI across the supply chain 113 Use of AI in logistics 164 Use of AI in trade finance 195 Use of AI in c

5、ustoms and trade compliance 226 Cost,complexities and payoffs of AI adoption 25The future of trade:collaboration for convergence 31Contributors 33Endnotes 35Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech2ForewordArtificial intelligence(AI)is not just transforming

6、global trade;it is revolutionizing it at an astonishing pace.AI is the driving force behind unprecedented levels of efficiency,sustainability and inclusivity,setting a new benchmark for global trade.Building on the foundation laid by last years comprehensive TradeTech report,which offered a broad lo

7、ok at technologys role in transforming global commerce,this edition delves deeper into the profound impact of AI across supply chains,trade finance,logistics,customs and compliance.AI has transitioned from aconcept to a tangible reality poised to revolutionize trade by streamlining processes,reducin

8、g costs and facilitating a more accessible and resilient global commerce ecosystem.This report underscores that realizing AIs full potential demands overcoming challenges that transcend borders and institutions.Establishing interoperable systems,ensuring verifiable processes,and aligning global regu

9、lations and incentives require collaborative,innovative solutions.Unlocking these opportunities will depend on forging partnerships and setting shared standards to ensure the benefits of AI-driven trade extend equitably across the globe.The United Arab Emirates and the World Economic Forum are pleas

10、ed to partner on the TradeTech Global Initiative,aiming to revolutionize global trade.The United Arab Emirates is rapidly emerging as a global powerhouse and a central hub for innovation in trade.Together,we are committed to empowering businesses and communities worldwide,ensuring noone is left behi

11、nd in this era of transformation.Let us unite in shaping the future of trade,where AI catalyses seamless cross-border commerce,shared prosperity and inclusive opportunities for all.We call upon all stakeholders to join us in propelling this bold vision forward.Brge Brende President and CEO,World Eco

12、nomic ForumArtificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTechJanuary 2025Ahmed Jasim Al Zaabi Chairman,Abu Dhabi Department of Economic DevelopmentThani bin Ahmed Al Zeyoudi Minister of State for Foreign Trade,Ministry of Economy of the United Arab EmiratesArtificial

13、Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech3Executive summaryFrom enabling small firms to compete globally to making supply chains more sustainable,artificial intelligences(AI)impact extends far beyond efficiency gains.This report examines both how AI is currently being u

14、sed in trade and the choices that will define its future effects.Concrete examples of AI use in supply chains,logistics,trade finance,and customs and compliance are presented throughout the report,demonstrating the range of benefits AI can unlock in terms of efficiency,sustainability and inclusivity

15、.These include:Opening new markets,particularly for small-andmedium-sized enterprises(SMEs)Enabling efficient automation Enhancing supply-chain visibility Promoting sustainability by supporting circular economy models and targeting carbon emissions for reduction Strengthening the resilience and secu

16、rity ofglobal trade However,these benefits may not extend throughout the entire trade ecosystem,depending on the extent to which AI is implemented.The successful implementation of AI across globaltrade,or“AI convergence”,can significantly increase real trade growth while enabling efficiency,sustaina

17、bility and inclusivity gains across the tradeecosystem.The fragmented implementation of AI across global trade,or“AI divergence”,can increase real trade growth but concentrate the benefits of AI into“trade islands”AI-powered hubs that offer superior efficiency but with limited participation.This cou

18、ld exacerbate global inequality in the short term,much as containerization initially benefited North-North trade more than other flows.Despite the incentives to use AI in trade processes,the difficulty of implementation threatens ecosystem-wide adoption.This report offers an actionable matrix,indica

19、ting the specific areas where AI adoption is more likely given its potential impact relative to its ease of implementation,highlighting areas where stakeholder incentives are aligned and areas where more extensive collaboration is required.Four factors can increase the ease of AI implementation in t

20、rade:1.Ensuring system interoperability2.Building trust 3.Facilitating public-private partnerships(PPPs)toalign incentives4.Investing in workforce development and digitalinfrastructure to strengthen human-AI collaborationAchieving both rapid innovation and inclusive participation requires extensive

21、collaboration and coordination across all parties in the trade ecosystem.Thankfully,AI can facilitate just that.AI is transforming global trade in ways that reshape both who participates and how business is done.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech4Intro

22、duction:AIs potential as atrade technologyArtificial intelligence(AI)could contribute$19.9 trillion to the global economy by 2030,with the potential to reshape international trade dynamics.1 This report,Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech,maps the most

23、dynamic areas of AI-driven change in trade.By examining concrete examples across supply chains,logistics,trade finance and customs,it evaluates how AI adoption can create a more efficient,sustainable and inclusive trading system.AI technology offers extraordinary possibilities.It enables trade parti

24、cipants to coordinate at unprecedented scales and opens new ways for businesses,regulators and entrepreneurs to solve complex challenges together.Some experts believe that once organizations fully embrace AIs potential to transform work practices,including through reskilling their labour,there could

25、 be a significant boost in global productivity.2Fully embracing the potential of AI across the trade ecosystem can result in AI convergence,under which cumulative real trade growth in goods and services would increase by 13.6 percentage points,with uniform benefit across economies.AI divergence,or a

26、 fragmented implementation of AI,would only increase cumulative real trade growth by 9.3 percentage points and would concentrate gains,exacerbating inequality.3 AI divergence could lead to the creation of“trade islands”AI-driven centres where trade flows become concentrated due to the enhanced effic

27、iency and cost savings enabled by AI.AI is expected to add$19.9 trillion to the global economy by 2030,inevitably impacting global trade.13.6percentage pointsCumulative real trade growth in goods and services under the successful implementation of AI in trade5Artificial Intelligence for Efficiency,S

28、ustainability and Inclusivity in TradeTechAI adoption builds on trends that are already under way:businesses are increasingly treating data as a product,improving quality and usability,and shifting from proprietary technology towards open,flexible solutions.These changes create a positive feedback l

29、oop,encouraging wider adoption of collaborative systems.The complexity of coordinating todays supply chains and the blended sales of services and goods create a market ripe for AI.Already,services account for between a third and a half of the value added in manufacturing exports,including,for exampl

30、e,research,product development and after-sales support.4 The increasing relevance of services in global trade amplifies the potential value of AI in the creation and delivery of these embedded services.AI innovations are thriving in highly tradeable sectors,and those sectors are in turn investing he

31、avily in AI.More than 90%of AI-related patents are concentrated in five sectors:computer and electronics,machinery,IT services,transport equipment and electrical equipment.5 These factors combine to create a snowball effect,with AIs role inglobal trade set to expand dramatically.An important conside

32、ration in the adoption of technology is the difficulty of introducing it.In the World Customs Organizations(WCO)Smart Customs Survey,customs officials rated the incorporation of technologies such as virtual reality and blockchain as difficult,whereas introducing AI was rated as relatively easier(see

33、 Figure 2).Moreover,the potential impacts of AI adoption were seen as considerably higher than other technologies,such as 3D printing.Given limited resources for frontier technologies,a technology that is relatively easy to implement,with high expected impact,is set for high growth.AI adoptionThe AI

34、 avalanche in tradeFIGURE 1Increased growth in services and servicification further increases Al opportunities in tradeNew and efficient services further push the servicification trend in supply chains,increasing growth in services1234Growth in services as a result of Information and Communication T

35、echnologies,especially digital services,increases cross-border data flowsAl can enhance the delivery and efficiency of services,while also creating new services and business opportunities The increasing relevance of services in global trade amplifies the potential value of AI in the creation and del

36、ivery of these embedded services.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech6Expected impact vs.ease of adoption of different technologies in customsFIGURE 21.02.03.04.09.05.00Ease of adoption(average)5.06.07.08.0Impact(average)4.0RobotsVirtual reality(VR)3D pr

37、intingDronesBiometricsBlockchainBig data and data analyticsGeospatial mappingInternet of things(IoT)Cloud computingAISource:Based on World Customs Organization(WCO).(2024).Smart Customs Project:Results of the WCO Smart Customs Survey.AI goals in tradeMaqta Technologies Groups AI-powered Trade Single

38、 Window innovation BOX 1Developed and operated by AD Ports Groups Maqta Technologies Group under the supervision of the Abu Dhabi Department of Economic Development(ADDED),the AI-optimized Trade Single Window(TSW)system uses a multimodal approach to create a trade ecosystem that integrates air,land

39、and sea operations.TSWs traditionally allow traders to submit all necessary documents required for trading into a single platform.The AI-powered Advanced Trade and Logistics Platform(ATLP)goes beyond that to facilitate inspection,customs clearance and connections to economic zones.ATLP integrates nu

40、merous data sources to enable real-time tracking of goods,predictive analytics for supply chain optimization and streamlined trade processes.The project creates a paperless channel that fuses information sources covering 176 million digital transactions.ATLPs data repository draws on over 20 maritim

41、e,aviation,rail,regulatory,financial and economic zone sources.Further,Maqta Technologies Group provides robust cybersecurity governance and ISO-certified protocols to ensure reliability.These security provisions create a foundation for testing and refining regulatory approaches and supporting the d

42、evelopment ofnew technologies for global deployment.Since its implementation,ATLP has reduced documentation requirements for traders by 90%and cut the burden on workers time by 50-90%.Source:Maqta Technologies GroupEfficiency AI is adept at synthesizing statistical information to enable optimization

43、 and error reduction.Deploying AI effectively can free up human labour traditionally required to perform routine or dangerous tasks and allow workers to focus on higher-value,morecomplex activities.By automating and broadly optimizing trade processes,AI can supportgreaterefficiency throughout the en

44、tire supply chain.As AI adoption in trade increases,the technology is showing early gains in three goals:efficiency,sustainability and inclusivity.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech7Sustainability Identifying and quantifying environmental impact repres

45、ents one of the most effective environmental levers.AI can provide new insights to address Scope 3 emissions,which refer to all the emissions produced across an organizations value chain and can account for 70%to 95%of an organizations environmental footprint.6 Individually assembling data on facili

46、ties is a complex task given that such properties operate independently from the reporting company,but AI tools are able to generate accurate estimates of emissions at each stage of the production process.This AI-driven approach can produce Scope 3 emissions estimates at scale with previously unatta

47、inable detail.AIs ability to improve supply chain visibility makes it invaluable for assessing environmental footprints,as required by the EUs Carbon Border Adjustment Mechanism,for example.Scope 3 emission measurement BOX 2Reducing Scope 3 emissions requires firms tohave visibility,tracking and mea

48、surement for each input along complex product supply chains.Measuring Scope 3 emissions for anautomobile,for example,requires tracking thesteel and aluminium production involved in itscomponents.AI and a range of new technologies can support efforts to reduce Scope 3 emissions by:Improving specifici

49、ty of emissions informationinputs Gathering facility-level data to improve accuracy Reducing labour and costs involved in Scope 3 measurements Improving collaboration with regulators and partner companies to accurately measure emissions and work to minimize them Protecting data privacy and sovereign

50、ty byusing federated learning modelsSuch tools can improve Scope 3 emission reductions to create better environmental outcomes,sanctions enforcement,compliance with trade regulations and overall efficiency.Source:AltanaAIInclusivity AI can support small and medium-sized enterprises(SMEs),which often

51、 lack the resources to navigate the regulatory complexity of global trade.Beyond the language barrier,challenges include finalizing contracts,obtaining finance and insurance,securing raw materials,and handling payments at each step of the process.Complex regulatory environments exacerbate the risks

52、of non-compliance and create barriers to cross-border trade.Automation and support through AI can alleviate theburdens on SMEs and create a more level playing field by:Breaking down language barriers to grant access to global markets Helping businesses better comply with regulations Anticipating shi

53、pping bottlenecks and supply chain backlogs Predicting and suggesting responses to shocksDHL MyGTSDHL offers a self-serve platform,MyGTS,that enables shippers to compare and analyse trade regulations and requirements between the exporting and importing country,highlighting different regulations and

54、requirements for goods to travel from export location to final customer.Thetool helps companies plan their expansion into new markets and develop strategies to maximize their competitive advantages.MyGTS uses AI to distil trade agreements and estimated duties and taxes for sending various goods acro

55、ss different trade routes.The toolswhich include estimates for shipping fees,insurance,customs duties and cross-border taxes improve businesses ability to estimate the total cost of transporting products to their final destinations.Source:DHL BOX 38Artificial Intelligence for Efficiency,Sustainabili

56、ty and Inclusivity in TradeTechAI:A composite technology for trade1AI tools integrate long-standing statistical and computer science methodologies into digital technology,drawing on decades of technological developments.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTe

57、ch9Predictive AI relies on statistical analysis and machine learning(ML)to find patterns in data,making it particularly suitable for trend analysis and forecasting.Generative AI(genAI)draws on complex neural networks such as large language models(LLMs)and is especially powerful for activities such a

58、s generating and translating text,processing images and planning tasks.7 Many of todays most powerful AI applications in trade draw on a combination of these models to produce decision support and automated insights for harmonizing data,predicting trends and more.Typologies of AI and their applicati

59、ons in tradeBOX 4Predictive AI builds on probabilistic models(including Bayesian statistics)and ML to,forexample:Fill information gaps in trade statistics anddocuments based on historical data Optimize warehouse inventory Assess risk and develop risk profiles for traders Authenticate information and

60、 detect counterfeits and fraud Forecast economic trajectories with increasingaccuracyGenAI draws on complex neural networks like LLMs or generative adversarial networks(GANs)and ML to,for example:Generate and translate text Simulate complex scenarios to develop contingency plans Respond to queries f

61、rom clients and traders Summarize information from trade documents Generate trade documents,such as customsdeclarations Most TradeTech tools are compound AI systems,which combine AI types together and can be paired with other innovations,such as breakthroughs in big data and robotics(see Figure 3).F

62、or instance,genAI tools can help create an invoice or bill of lading,but to do this accurately often relies on data harmonized and processed through other AI techniques,which constitute the most important applications of AI in trade.The term“AI”in this report encompasses all instances of AI usage,in

63、cluding compound AI systems and their pairings with other technologies.Intersection of AI and advanced technologiesFIGURE 3Artificialintelligence(AI)Big dataPredictive analyticsCloud computingAl as a serviceAutonomousvehicles and drones5GSmart contractsBlockchainAutomatedwarehousesRoboticsPredictive

64、 maintenanceand logisticsInternet of thingsArtificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech10Use of AI across the supply chain2Todays supply chains enable a staggering degree of complexity,which AI can simplify.Artificial Intelligence for Efficiency,Sustainability a

65、nd Inclusivity in TradeTech11The mineral inputs for batteries of electric vehicles(EVs)travel roughly twice the circumference of the earth from their origin to production,on average.8 AI has the potential to upend these complex global supply chains,transforming nearly every component of modern suppl

66、y chains and changing how the systems parts interrelate.Thus far,its impact has primarily been a story of the private sectors use of reliability-and efficiency-orientated tools,as well as its use of AI to reduce risk and support sustainability efforts.The use of AI in supply-chain optimization has r

67、aised service levels by 65%and reduced logistics costs by 15%,according to one recent study.9At the ecosystem level,AI can also improve decision-making in both the public and private sectors by presenting more sophisticated tools to visualize and analyse the current state of supply chains.Historical

68、ly,information challenges,from tracing the provenance of goods to modelling weather disruptions,have limited companies ability to reliably manage supply chains.By improving visibility and data coordination,AI can provide a clearer picture of todays complex and interconnected supply chains.Informatio

69、n verificationTodays supply chains require companies to manage compliance,procurement,resiliency,economic security,environmental harm mitigation and efficiency across long distances and to do so at scale.Historically,supply chain and life cycle information has been fragmented and opaque.AIhelps stre

70、amline and integrate real-time data on supply chains into overall decision-making,which helps institutions anticipate rapid change.AI can fuse contextual information found in corporate registries,transport documents,purchase orders and related records with language processing across all major langua

71、ges to create new insights about modern supply chains.Fusing data sources enables the production of continuously updated maps,which provide new levels of visibility about complex,multi-tiered supply chains.Even relatively simple supply chains like the one in Figure 4 can involve various companies ac

72、ross multiple tiers and jurisdictions.Charting supply chainsFIGURE 4Buyers tier oneProductTier zero Suppliers tier oneSuppliers tier twoSuppliers tier threeBuyerstier twoCompany CCompany AVehicle gear boxesAutomotivecompanyCompany ACompany CCompany ECompany BCompany DCompany FCompany GCompany BCompa

73、ny DCompany ECompany FSource:Altana AI65%increase in service levels as a result of theuse of AI in supply-chain optimization.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech12Digital ID can help to build data reliability for AI-based decision-makingBOX 5Digital ID f

74、acilitates deeper implementation of AI throughout supply chains and is especially important to enable autonomous trade,under which AI agents automatically execute transactions based on predetermined criteria.Immutable IDs that parties can trust can be built using technologies such as the blockchain

75、or cryptographic signatures,which AI can use to inform its decisions.Given the complexity of global supply chains,verifying the authenticity of legal entities,inputs,products and documentation has become increasingly difficult.AI can offset these challenges by automatically authenticating all partie

76、s in a supply chain using a single,global“root of trust”for identities,which offers a neutral source of truththat attests to the authority of counterparties.Source:Key State Capital Supply chains for humanitarian needsBOX 6Humanitarian supply chains are vital for delivering life-saving goods to vuln

77、erable communities,particularly in crises.Emerging technologies,including AI,have the potential to strengthen these supply chains.10 Tools under development,such as the Global Supply Chain Dashboard(GSSD),aim to provide real-time visibility into potential disruptions like security incidents,port con

78、gestion and natural disasters.This enhanced visibility could help humanitarian relief organizations navigate logistical challenges and ensure the delivery of critical resources,such as therapeutic food for malnourished children.By using anonymized data from public and private sources,AI-driven predi

79、ctive analytics can anticipate risks,suggest alternative routes and prioritize resources.Improved data sharing and visibility promise greater resilience,enabling more coordinated and timely responses when supply chain shocks occur,ultimately saving lives.AI-powered mapping provides an information ba

80、seline that supports a broad range of other tasks,such as:Compliance:Verifying inputs improves companies ability to ensure compliance across multiple regulatory environments.Resilience:Improved supply chain mapping allows businesses to anticipate and improve responses to potential disruptions or bot

81、tlenecks(see Box 6).Insurance:Better data mapping and analysis improves participants understanding of their concentration risks and ability to develop contingency plans for business interruptions.Optimization of supply chains:Better visualization and mapping of supply chains can improve the way orga

82、nizations optimize their production inputs,through reducing transport inefficiencies and streamlining supplier relationships,among other improvements.Another approach to verification involves using AI-enabled track-and-trace systems that monitor goods from production to sale.The layering of ML and A

83、I-powered maps can analyse patterns toimprove tracking and identify risks of illicit trade.Better visualization and mapping ofsupply chains can improve the way organizations optimize their production inputs.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech13Predictin

84、g tomorrows repairs,todayBOX 7Time series data from sensors are used to help predict when equipment will fail and when maintenance is necessary.GenAI uses new technology to analyse textual information and overcome gaps in data to better understand machine conditions and failure modes.The tools provi

85、de more accurate maintenance schedules that integrate into existing business systems.Data analysis systems can create a repository ofknowledge based on past cases to ensure the preservation of institutional knowledge,regardless of the experience level of individual technicians.These models can incor

86、porate data in multiple languages,which enhances businesses ability to develop wide repositories of knowledge about their equipment andreduce product downtime and failures.Transforming waste into worthBOX 8AI and robotics can change recycling for EV car batteries.SK Ecoplant aims to use AI tools to

87、determine the location of the battery and bolts,the protocols for extracting the unit and the condition of the parts,including information about broken or missing pieces,enabling the automation of battery pack removal.This information improves the operation of robotics,which are used to unscrew bolt

88、s and remove the covers of battery packs.The combination of AI and robotics can increase safety and decrease the processing time for dismantling battery modules by 32%.The company also expects to implement AI and automation to extract battery module cells by 2026,with the goal of improving overall e

89、fficiency of recycling operations by 30%.Source:NUS Business SchoolFraud detection The value of counterfeit and pirated products isestimated at$464 billion,constituting 2.5%of world trade.AI is accelerating the transition from linear to circular economy models,which prioritize resource conservation,

90、reuse and waste minimization.Improving supply chain analysis represents one strategy for companies to improve their circularity and strengthen sustainability efforts.AI and blockchain enable this transition by providing the in-depth analytics and visualization required to track product life cycles a

91、nd identify opportunities for reuse and recycling,promoting transparency,traceability and data-driven decision-making.Concrete applications of AI for circularity include:Product life cycle extension:AI-powered predictive maintenance and data analytics can extend the lifespan of various goods,ranging

92、from manufacturing equipment to consumer electronics,that can reduce waste(see Box 7).Waste management:AI and robotics are automating and improving e-waste recycling,especially for minerals in electric batteries,by increasing efficiency and reducing harm to humans(see Box 8).Emissions tracking:Monit

93、oring Scope 3 emissions allows the environmental impact of a good to be quantified,allowing for more informed decision-making(see Box 2).The growth of electronic commerce(e-commerce)has introduced vast new opportunities for fraud,such as fake reviews,counterfeit products and scams.Estimates suggest

94、that counterfeit and pirated products constitute 2.5%of world trade,or$464 billion in value.11 AIs data analysis capabilities can flag suspicious transactions.Amazon stopped over 700,000 new accounts from bad actors and identified over 7 million counterfeit products around the world in 2023.12 Simil

95、arly,Alibaba identified$159 million worth of counterfeit products between July 2022 and June 2023 and supported law enforcement operations that resulted in 619 arrests and 257 facility closures.13 Fraud detection services also present business opportunities for emerging companies such as Sift,Entrup

96、y and Signifyd,which have also capitalized on technological innovation.While studies suggest an increase in counterfeit products infiltration of supply chains,AI can support the detection of sophisticated fakes.14AI for a circular economyArtificial Intelligence for Efficiency,Sustainability and Incl

97、usivity in TradeTech14Smart sourcingAIs ability to analyse complex,multidimensional data in real time can revolutionize the sourcing and movement of critical raw materials.In the past 20 years,trade in energy-related minerals has risen from$53 billion to$378 billion a compound annual growth rate of

98、10%,which is likely to increase as the worlds reliance on advanced technologies grows.15 But the historic volatility of metal prices poses challenges for consumers who need stable prices and reliable supply chains.AI can help offset the volatility of these markets through:Rapid and incisive analysis

99、:Price data,globaleconomic statistics and market sentiment are all factors that affect supply chains.AI and especially ML tools can identify patterns that enable leaders to make more accurate predictions about future pricing.Real-time insights:AI tools can continuously monitor news and social media

100、to assess how various events may impact metal prices.Thisprovides real-time insights that enable faster decision-making.Enhanced risk management:AI can easily identify unusual trading patterns or price movements,allowing for quicker detection of market inefficiencies and new trading opportunities.Pa

101、ttern recognition can flag errors and reduce risks by identifying trades that do not fit expected behaviour.Faster trade execution:AI-driven systems can automate the process of capturing and entering trades into trading platforms,reducing manual input errors.The assets of AI can augment human intell

102、igence and create a smarter,better-informed workforce with greater visibility over global markets.By reducing risk and costs,such tools allow access toa greater supply of critical raw materials at better prices for end consumers.Current and future applications of AI across supply chainsFIGURE 5Curre

103、nt applications:Supply chain mappingParty authentication through digital IDInternet of things(IoT)sensor networksFlagging suspicious transactionsand illicit tradeCounterfeit detectionFake review detectionFuture applications:Full traceability through transparentsupply chainsCurrent applications:Facil

104、ity-level Scope 3 emissionsmeasurementPredictive maintenanceDemand predictionComputer vision for automated recyclingFuture applications:Global Scope 3 emissions trackingMaintenance automationZero waste in manufacturingPerfect demand prediction,eliminatingsurplus inventoryDe-centralized autonomousman

105、ufacturing and delivery units(AMDU)Current applications:Demand predictionMarket history analysisFraud alertsTranslation and localizationDynamic pricingFuture applications:De-centralized trade networksand peer-to-peer transactionsAI-managed decentralized autonomousorganizations(DAOs)AI agents trading

106、 automaticallythrough digital IDsFlexible factories and synthetic digital assetsEthical sourcing through transparentsupply chainsPerfect localizationTraceability,fraudand information verificationCircular economySourcing and purchasing The assets ofAI can augment human intelligence and create a smart

107、er,better-informed workforce with greater visibility over globalmarkets.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech15Use of AI in logistics3AI is transforming logistics,facilitating the delivery of new services.Artificial Intelligence for Efficiency,Sustainabil

108、ity and Inclusivity in TradeTech16Predictive analytics:AI excels in using voluminous data to predict outcomes of logistics operations.Such predictions can help firms anticipate cascading effects,such as the ripple effects of flooding in a rural province or a strike at a single port.AI can integrate

109、vast amounts of both real-time and historical data,such as:Trade trends Weather conditions Port traffic Geopolitical events Economic indicators Social media trendsAI can also generate new insights on interrelations,thereby helping logistics providers plan routes and forecast demand.Such predictive a

110、nalytics improve businesses ability to optimize inventory and reduce waste.Moreover,pairing AI with othertechnologieslike internet of things(IoT)improves firms ability to plan maintenance schedules and anticipate equipment failure(see Box7).By organizing timely repairs,firms can reduce unexpected do

111、wntime andextend assets lifespans.These analyses become more accurate over time as AI continuously learns from and adapts to past predictions and newdata.Digital twins and simulation:Modelling and simulations have long been a tool for logistics providers,and AI enhances these traditional tools by fe

112、eding real-time data into evolving simulations.The technique of creating“digital twins”allows logistics operators to replicate supply chain networks in digital models to represent ports,warehouses,distribution centres and other transit nodes.By incorporating sensors,IoT and other data sources,digita

113、l twins allow logistics operators to manage potential disruptions proactively.Designing and simulating supply chains rather than just accepting inheriting real supply chains gives users more flexibility to craft intelligent responses.These analyses become more accurate over time as AI continuously l

114、earns from and adapts to past predictions and new data.Digital twins by Maersk BOX 9Maersk has adopted AI in pricing,route optimization and forecasting models.Its digital twin modelling allows port operators to assess responses if storage facilities approach capacity,strikes loom or weather changes.

115、Digital twins,powered by data from sensors,can optimize how the operator should plan the loading and unloading of containers under such conditions.AI-powered simulations enable the analysis of historical and current data to maximize efficiency.Moreover,such tools increase overall safety by improving

116、 training.Port workers and vessel crews can use digital twins to simulate high-risk scenarios,thereby reducing accidents and improving resilience.Source:Maersk17Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTechCurrent and future applications of AI in logisticsFIGURE

117、6Current applications:Digital twinsDemand prediction for inventory allocationRoute optimizationPredictive maintenanceSupply chain resilience testingIdentification of bottlenecksFuture applications:Autonomous decision-makingDynamic reroutingAdaptive live restructuring of supply chainsOptimized invent

118、ory levels across networksMitigation of bottlenecksCurrent applications:Real-time trackingCondition monitoringETA predictionFacility-level Scope 3 emissionmeasurement Future applications:Real-time shipment visibility Augmented smart containers(IoT)Global Scope 3 emission measurement Current applicat

119、ions:Autonomous vessels Smart warehouses Customer service chatbots Future applications:Autonomous fleets for seamless delivery Connected smart warehouses Intelligent virtual assistants Predictive analyticsand simulationEnd-to-end visibilityAutonomous operations End-to-end visibility:AI is enhancing

120、supply chain transparency through real-time tracking and monitoring of shipments.By integrating AI with GPS and IoT sensors,logistics providers can determine the exact locations and conditions of goods,reducing transit losses.Improved end-to-end visibility can also pave the way for new services,rang

121、ing from trade compliance to supply chain resilience testing.Automation and autonomous operations:AIcan streamline routine tasks,from document automation to autonomous vessel navigation.Adopting automation in targeted sectors can reduce human error and free up human resources for more strategic acti

122、vities.Picking,packing and inventory management represent tasks that AI-powered robots have mastered to produce overall gains in efficiency and accuracy.An emerging frontier involves the use of AI to power chatbots and virtual assistants for routine inquiries and shipment updates.Additionally,AI is

123、enabling the development of self-navigating vessels that use a combination of sensors,GPS and ML algorithms to plot optimal routes,avoid collisions and adapt to changing weather conditions.Overall,the successful integration of AI in logistics will require a collaborative effort among all stakeholder

124、s in the supply chain,including shipping companies,technology providers,regulators and standard-setting bodies.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech18Use of AI in trade finance4AI is revolutionizing trade finance,reducing creditdecisioning times from week

125、s to minutes.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech19A total of 80%of trade involves some form of financing,and trade finance tools range from credit lines to payment guarantees to specialized insurance.16 Access to credit is especially critical for SMEs,w

126、hich might otherwise struggle to front the costs of cross-border trade.AI can work alongside other technologies to reducethe gap between the demand for and availability of trade finance a funding shortfall of$2.5 trillion in 2022.17 AI is offering improvements inthe following domains:Reducing ineffi

127、ciency:Reliance on paper-based documentation greatly increases processing delays.Lowering barriers to entry:Onboarding costs,transaction fees and other information gaps can make lenders cautious about relatively unknown SMEs.Decreasing fraud risk:Paper-based documentation,which is central to trade f

128、inance,makes the field vulnerable to fraud.Increasing transparency:Trade finance involves a complex array of intermediaries and processing steps,which impede transparency.One of the most socially significant contributions of AI in trade finance involves greater inclusivity,as lack of credit dispropo

129、rtionately hurts SMEs.Challenges faced by SMEs include a lack ofcollateral and long-term relationships with bankers,insufficient credit or transaction history,and unfavourable market conditions,resulting in 45%oftrade finance rejections being attributed to SMEs.18 AI can help level the playing field

130、 by providing new tools for lenders and connectivity with borrowers.AI can work alongside other technologies to reduce the gap between the demand for and availability of trade finance a funding shortfall of$2.5 trillion in 2022.Breaking trade finance barriers in AfricaBOX 10Jetstreams AI-enabled too

131、ls are facilitating access to trade finance for SMEs in West Africa,resulting in more than$8 million in loans and financial guarantees to hundreds of businesses in Nigeria and Ghana.This achieves a loss rate that is less than half the regional industry average.At the start,the JetAssist application

132、used AI to pull status updates from templated emails sent by global shipping lines and push them to a dedicated Slack channel to efficiently parse and relay critical shipping information,improving visibility and communication.Building on this using optical character recognition(OCR),data is now extr

133、acted from financial and logistics documents issued by third parties.It is then uploaded to a structured database to be fed into algorithms which help to cut down credit decisioning time from the industry-standard of around one month to less than one minute.Source:Jetstream AfricaSupply chain financ

134、e depends on in-depth understanding of the risk profile of a transaction,including the actors involved and their track records.AI helps assess the relevant factors to reach a financing decision.Further,the combination of AI and optical character recognition helps digitize the underlying invoices,shi

135、pping documents,customs filings and other documents exchanged between buyers and sellers.Digitalizing such data improves efficiency,reduces errors and accelerates exchanges.For example,Twinco Capital uses AI to extract information from unstructured shipment paperwork,reducing processing time by 80%a

136、nd increasing accuracy.20Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTechCurrent and future applications of AI in trade financeFIGURE 7Current applications:OCR for data extraction and digitizationThree-way purchase order matchingAutomated transaction processingSmart

137、 contractsFuture applications:Autonomous supply chainsAutonomous transaction processingAugmented smart contractswith instantaneous settlementCurrent applications:Fraud assessmentCredit scoringMarket trend analysisNatural language processing forcompliance checksFuture applications:Instant transaction

138、-specific insuranceAutomated risk assessmentand underwritingCurrent applications:Data-based approvalsFacilitated decentralized financeEnvironmental,social and governance(ESG)scoringEnvironmental impact verificationChatbotsFuture applications:AI-powered DAO focused on SME financingAugmented micro-fin

139、ancing options for SMEsDocument andpayment processingRisk managementInclusive andsustainable financeArtificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech21Use of AI in customs and trade compliance5AI is streamlining customs processes and facilitating increasingly difficu

140、lt trade compliance.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech22Automating document processing represents a robust area of AI incorporation.Traditional customs procedures require clerks to process large volumes of paperwork manually,which can lead to human err

141、ors and delays.For example,the global logistics provider DSV employs more than 2,500 clerks who handle over half a million declarations each month.AI transforms processing practices.AI tools can extract information about each shipment from unstructured documentation about goods to ensure that custom

142、s forms are accurate.This automation not only accelerates the clearance process but also enhances accuracy by minimizing manual errors,which can result in fees,incorrect taxation information and delays.Another area where AI has shown success is improved cargo scanning.As the volume and complexity of

143、 global trade continue to increase,customs authorities face mounting challenges in detecting and intercepting contraband and counterfeit goods.AI can support new forms ofdata analysis,pattern recognition and predictive modelling that improve detection of high-risk shipments andmake more informed dec

144、isions in real time.Automation not only accelerates the clearance process but also enhances accuracy by minimizing manual errors,which canresult in fees,incorrect taxation information anddelays.Tracking tigers,combatting contrabandBOX 11Project SEEKER represents an AI tool specifically designed to i

145、mprove customs operations.The technology can detect illegally trafficked wildlife concealed within luggage and cargo by alerting enforcement agencies when an illegal wildlife item(such as tiger bones or elephant ivory)is detected.Project SEEKER was trained on X-ray images from baggage scanners in He

146、athrow Airport,which can scan up to 250,000 bags in a day.Ithas an over 70%successful detection rate for certain high-value items and complements existing digital security operations without requiring overhaul or modifications.Using cloud technology,Project SEEKER can go beyond single-airport scanni

147、ng cargo by connecting data patterns to build a comprehensive map of illegal wildlife tracking around the world.Such insights can help regulators identify smuggling routes and improve efforts to combat illegal trafficking.Source:MicrosoftDP World CARGOES CustomsBOX 12DP Worlds CARGOES Customs uses A

148、I-basedtools to improve customs operations andstreamline digitalization.The technology usesAI for:Valuation:Cognitive document processing extracts information from supporting trade documents to analyse data mismatches compared to data available in thecustomsdeclaration,reducing the manual verificati

149、on of data.Classification:Natural language processing helps traders and customs brokers identify Harmonized System(HS)classification based on descriptions of the product.Risk assessment:Predictive models analyse patterns in historical data to help customs authorities resolve misclassification and un

150、dervaluation issues.Virtual assistants:Chatbots provide declaration and payment statuses,inspection booking information and HS classification services for traders and customs brokers alike.Source:DP World23Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTechHS complex c

151、odesBOX 13Todays traders must navigate over 10,000 trade restrictions and tariff provisions introduced in the last decade.The complexity of these regulations creates pressure on logistics providers and customs brokers to operate in an increasingly volatile,high-stakes regulatory landscape.AI offers

152、some good news for traders who need to comply with these regulations and ensure shipments conform to standardized trading codes known as HS codes.Such codes are the backbone for coordinating shipping in many contexts;however,they can also add complexity.AI can facilitate traders compliance with vari

153、ous requirements by supporting language processing and documentation requirements.For example,the technology provider AltanaAI can:Automate more than 80%of classifications Improve the accuracy of automated HS classification from below 50%to 80%or more Audit existing classifications to ensure complia

154、nce Provide up-to-date tariff dataSource:AltanaAIMore than just incremental improvements,these customs changes have the potential to transform practices.Traditionally,customs operations have operated on a transaction basis,wherein authorities selectively inspected products at the border.AI and new t

155、echnology enable the shift to a systems-based approach,wherein businesses can integrate their shipping and information-management systems with customs and tax platforms,with risk-based inspections at any point.21 So far,the gains from such technology have largely been concentrated in regions where i

156、nformation infrastructures are already well developed.International coordination is needed to ensure a broader range of jurisdictions and entities can benefit from the potential gains ofAI-linked improvements.Current and future applications of AI in customs and complianceFIGURE 8Document processingR

157、isk managementComplianceCurrent applications:OCR for data extraction and digitizationDigital form creationAI-powered TSWs for automated filingFuture applications:Comprehensive business-to-government(B2G)and government-to-government(G2G)TSW system connectionCurrent applications:Cargo scanning for con

158、traband detectionRisk prediction targeting high-risk shipmentsFuture applications:Augmented cargo scanning for automatedclearance and complianceRisk profiles and tiered system forautomated clearance of low-risk shipmentsCurrent applications:HS code and tariff classificationRepository of compliance m

159、easuresAutomated assistantsFuture applications:End-to-end traceability for complianceand preferential access(rules of origin)Augmented and adaptive automatedassistantsPerfect HS code and tariff classificationAI can also help traders and officials navigate new regulatory requirements to combat forced

160、 labour.For instance,US regulators designed the Uyghur Forced Labor Prevention Act19 to bar products from the Xinjiang Uyghur Autonomous Region in China due to forced labour concerns.US Customs and Border Protection use AI-powered solutions to enforce the ban accordingly.20AI tools can enable office

161、rs and analysts to gain insight into obfuscation techniques such as shifting corporate networks that companies use to conceal forced labour within their value chains.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech24Cost,complexities and payoffs of AI adoption6While

162、 the benefits of AI are clear,its adoption in trade processes is more complex.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech25The adoption of any new technology creates costs and complexities:staff must be trained,new hardware or software must be acquired,and inst

163、itutions must adapt to new practices.Despite the upfront challenges,the prospect of AI innovation offers numerous advantages,and todays businesses risk being left behind if they neglect the prospects of AI adoption.How should traders navigate AI opportunities?Where can AI implementation yield the gr

164、eatest gains?An expert survey,mirroring the considerations of customs officials in the WCO Smart Customs Survey(see Figure 2),sheds light on the rate at which AI adoption is likely to proceed for various uses,considering the incentive for change and the ease of implementation.For example,businesses

165、have strong incentives to adopt AI when the efficiency gains are obvious.However,if the technology is challenging to implement forexample,requiring intensive worker reskilling andexpensive new hardware then the advantages would be less obvious.Likewise,if regulatory barriers complicate technological

166、 change,the likelihood of adoption plummets.Multiple factors affect both the incentive for and theease of implementation of AI tools,including the scale of adoption.In general,small-scale adoption tends to be easier to achieve,but gains may bemore limited.For example,company-specific solutions,such

167、as dynamic pricing,are generally easier to implement than solutions that require supply chain alignment,such as risk-management techniques.Harder still are the changes that require ecosystem collaboration,such as comprehensive primary data sharing.In the short term,the AI solutions likely to receive

168、 the greatest traction are those that offer company-level impacts and defined payoffs.Staged or incremental adoption can support the uptake of AI in more complex fields.However,relying only on local optimization could result in uneven uptake and missed opportunities to support better societal outcom

169、es.Coordination between regulators,industry consortia and neutral platforms could better support AI adoption forgreater societal gains by aligning incentives.Several priorities emerge as topics that benefit fromcoordinated AI approaches.Ease of implementation and level of incentive of AI use in trad

170、eTABLE 1Ease of implementation Low Medium HighLevel of incentive LowIntelligence and negotiations:AI can facilitate both data analysis and more informed negotiation practices.Examples of uses include using AI to process and analyse import and export records,aswell as modelling outcomes in multi-part

171、ynegotiations.MediumContracting and payments:AIcan analyse transaction data,contract terms,regulations and thresholds for triggering approvals and payments of smart contracts,greatly reducing the time needed for trade finance to be released.Risk management and resilience:New technologies can improve

172、 traders mapping of supply chains,analyses of risk horizons and the development of contingency plans.Monitoring severe weather patterns and port congestion,for example,and combining these inputs with fleet-based data can improve traders understanding of risksthroughout the supply chain.HighTrade opt

173、imization:AI can support traders through work streams such as calibrating inventory to better align with demand predictions and monitoring inventories and shipments in real time.Moreover,AI planning to enhance maintenance timing can reducedowntimes,while simulations help prepare contingency plans.Ma

174、rket facilitation:AIs linguistic capacity and knowledge gathering abilities can support businesses operating in a wide range of contexts bytranslating communications,adaptingadvertising and connecting withdiverse customers.Ethics and ESG reporting:AI tools can monitor supply chains to track material

175、s and products and assess suppliers to avoid unethical labour,illegal activities and environmental damage to more accurately score risk and comply with ESGrequirements.Border clearance:AI tools can help with a range of challenges related to complying with customs and border-crossing protocols,such a

176、s managing documentation requirements and scanning X-ray images of cargo forfraudand smuggling detection.Despite the upfront challenges,the prospect ofAI innovation offers numerous advantages,and todays business risk being left behind if they neglect the prospects ofAIadoption.26Artificial Intellige

177、nce for Efficiency,Sustainability and Inclusivity in TradeTechInteroperabilityMany institutions still rely on legacy systems that struggle to interface with modern business operations.Lack of resources,antiquated infrastructure and outdated regulatory frameworks can also slow theadoption of promisin

178、g technologies.Interoperability helps ensure that actors from different jurisdictions and regions can benefit from the shared gains of trade.Even if different platforms emerge,aligning protocols can help ensure that legacy systems remain interoperable.Such measures support the inclusion of numerous

179、traders,even when institutions move at varying speeds in incorporating new technologies.Developing common standards can also safeguard interoperability and facilitate the adoption of AI technologies in trade.Areas that benefit from common standards include:Communication and data protocols,which ensu

180、re high-quality data inputs,facilitate data sharing and enable integration with other technologies like IoT,help AI generate accurate and reliable insights,particularly in complex environments like global supply chains.Cybersecurity benchmarks,which mitigate the risks associated with increased digit

181、alization and AI deployment,ensure transparency and trust inAI systems.Standard definitions,such as what qualifies as a legally negotiable contract or definitions for“waste materials”,can streamline collaboration.The coordination of standards:Allows for more consistent adoption of AI through supply

182、chains Facilitates operations across jurisdictions forlogistics providers Streamlines customs processes Accelerates trade financeAt the same time,AI can be part of the solution by breaking down language barriers to enable communication across diverse cultures and markets,enabling the development of

183、consistent standards,regardless of the language of origin.TrustVerifiability is a cornerstone of both trust and compliance in international trade.Given the complexity of supply chains and the number ofinputs involved,parties need assurances that the information they receive(such as documents and ide

184、ntifiers)is correct.Digital ID enables these assurances for humans and AI alike.Verifiability represents a foundational challenge for AI systems,given that the quality of the underlying data is decisive in determining the quality of analysis that AI generates.The concept of“garbage in,garbage out”ho

185、lds true for AI systems.For AI tools to deliver on efficiency,sustainability and inclusivity,they must be trained on accurate and relevant data.As such,digital ID can help ensure the accuracy of data,improving the ability to generate high-quality datasets for AI training.However,while AI can assist

186、in processing and tracking information at unprecedented scales,traders must coordinate ways to ensure verifiability without compromising sensitive information.Technologies capable of graduated disclosure,such as zero-knowledge proofs(ZKP)and authentic chained data containers(ACDC),are emerging as po

187、tential solutions.AI can be part of the solution by breaking down language barriers to enable communication across diverse cultures and markets.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech27Innovation frontier:zero-knowledge proofs(ZKP)BOX 15ZKP represent a brea

188、kthrough in cryptographic technology that could transform how trade information is verified and secured.This emerging technology allows parties to verify trade transactions and compliance without exposing sensitive underlying data.When combined with blockchain and AI applications,ZKP offer two keyad

189、vantages:Enhanced privacy:Enables verification of trade compliance while keeping transaction details confidential Reduced concentration risk:Increases system resilience through decentralizationFor example,customs agencies could verify regulatory compliance without accessing complete transaction deta

190、ils,effectively balancing transparency requirements with data privacy needs.Despite its transformative potential,ZKP technology remains in its infancy.One challenge involves the need to develop governance frameworks for ZKP implementation.Regulatory sandboxes offer a promising approach to accelerati

191、ng responsible adoption while developing appropriate governance structures.Source:Key State CapitalAssurance and accountability through digital IDBOX 14The Global Legal Entity Identifier Foundation(GLEIF),a non-profit organization established by the Financial Stability Board,created a global digital

192、 ID system known as the verifiable Legal Entity Identifier(vLEI).The system creates unique identifiers or cryptographic signatures for every transaction,supplier and purchaser.These identifiers cannot be altered by downstream participants in the supply chain and can,therefore,help reduce counterpart

193、y risk.vLEIs can be used to cryptographically sign records at every point in trade transactions,thus they enable verifiable ways to establish reputations and help reduce the scope for fraud and data processing errors.While the vLEI system does not completely eliminate all risks for fraud,it nonethel

194、ess offers a mechanism for establishing accountability previously impossible in digital systems.Such a foundation can facilitate real-time adaptation and defensibility of supply chains.Source:Key State CapitalPublic-private partnershipsImplementing AI across global trade demands coordinated action,n

195、ot isolated initiatives.Thischallenge is complex:supply chains span hundreds of entities with competing interests,varying resources and different capabilities to adoptnew technology.While effective AI systems require comprehensive data sharing,current structures struggle to balance collaboration wit

196、h competitive interests.Public-private partnerships(PPPs)offer the mostviable framework for driving ecosystem-wide AI adoption.By combining private innovation with public-sector support,PPPs can ensure AI implementation is technically sound,sustainable and inclusive.These partnerships can operate at

197、 multiple scales,from modernizing individual ports to developing cross-border frameworks that help businesses adapt to technological change.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech28Distribution of duties to maximize PPPsFIGURE 9Coordinate globaldevelopment

198、andimplementation effortsDevelop internationalstandards and guidelines Provide platformsfor multistakeholderdialogue Accelerate capacitybuilding in developingcountriesEnsure transparencyIncentivize participation Enable digitalinfrastructure andinnovation with theright building blocks Develop workfor

199、ceskills and planfor innovation ona national scale Provide fundingand resources Align national strategiesand priorities Ensure ethical andresponsible use of AI Facilitate cross-border collaborationon AI projectsProvide technologicalexpertise and innovation Invest in researchand development Provide f

200、eedback oncurrent bottlenecksand roadblocks Test and implementAI solutions in practicalbusiness contexts Offer operationalefficiency andmanagement expertiseConduct theoretical andapplied researchReskill the currentgeneration and preparefuture generationsContribute toethical frameworks for AI develop

201、mentRepresent publicinterests by monitoringand raising awarenessabout societal impactsFacilitate publicdialogue andengagementGovernmentsBusinessesAcademiaCivil societyInternationalorganizationsGovernmentsBusinessesAcademiaCivil societyInternationalorganizationsNote:No one party can generate this bro

202、ad range of contributions and insights;therefore,collaboration is the only way to maximize the benefits of AI adoption in global trade.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech29TradeTech Regulatory SandboxBOX 16The TradeTech Regulatory Sandbox aims to bridge

203、 the gap between cutting-edge technological solutions and existing local regulatory frameworks by creating a controlled environment in which start-ups,financial institutions and regulators can test new technologies in real-world conditions.Structured as a collaborative initiative,the sandbox require

204、s active engagement from both public and private sectors;government agencies,including the Central Bank of the United Arab Emirates,Abu Dhabi Global Markets(ADGM),Dubai Financial Services Authority(DFSA)and RegLab,provide regulatory oversight,while private sector participants bring technical experti

205、se and innovation.The sandbox currently focuses on trade finance,emphasizing AI and blockchain to make trade finance processes more efficient,sustainable and equitable.Although the initial focus is sector-specific,the sandbox model is adaptable and could expand to other areas in the future.While thi

206、s cohort is headquartered in the United Arab Emirates and managed by the World Economic Forum,the TradeTech Regulatory Sandbox has a global reach,drawing businesses from around the world.The insights and outcomes are not confined to United Arab Emirates markets but aim to inform international regula

207、tory best practices,positioning the sandbox as a pioneering initiative with the potential for broad cross-border impact.The Trade Worldwide Information Network(TWIN)exemplifies a targeted cross-border approach;it focuses specifically on the secure storing and sharing of trade-related information thr

208、ough distributed ledger technology.This can be augmented by AIs ability to digitize documentation,further facilitating information sharing.By bringing together governments,businesses,and non-governmental organizations(NGOs),the project demonstrates how focused collaboration can dramatically reduce t

209、ransaction costs and processing times,particularly benefiting SMEs.The complexity of global trade demands comprehensive approaches that span both jurisdictions and issues.The TradeTech Global Initiative,a partnership between the World Economic Forum and the United Arab Emirates,represents this new g

210、eneration of holistic PPPs.The initiative operates on several levels and scales.It facilitates local projects,such as the regulatory sandboxes within the UAE,that can be scaled to support other jurisdictions.The project also maintains a global perspective through regular convenings with internationa

211、l leaders and experts.By combining local implementation with global knowledge sharing,the TradeTech Global Initiative creates a powerful mechanism to harness successful AI solutions to benefit the whole trade ecosystem.The project creates a forum to test innovations in specific contexts while mainta

212、ining a platform for global dissemination.As such,it can accelerate the adoption of effective AI solutions and provide a blueprint for new models of public-private collaboration that transform trade.The TradeTech Global Initiative can accelerate the adoption of effective AI solutions and provide a b

213、lueprint for new models of public-private collaboration that transform trade.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech30The future of trade:collaboration for convergenceThe successful implementation of AI stands at a critical juncture,with two paths ahead:AI

214、convergence and AI divergence.Convergence could boost real growth in global goods and services trade by 13.6 percentage points through 2040,while divergence could yield a more modest increase of 9.3 percentage points a difference ofapproximately 38%.22AI divergence trade islands AI divergence,result

215、ing from an asymmetric and fragmented implementation of the technology,could lead to“trade islands”,wherein trade flows are concentrated through countries that quickly and fully adopt AI as a result of the increased efficiency,transparency and traceability possible through this technology.History of

216、fers a telling parallel incontainerization,which increased North-North trade by approximately 700%over two decades,while North-South and South-South trade saw substantially lower gains.23 This concentration need not be permanent:containerization eventually became the global standard,and late adopter

217、s benefitted from the lessons learned and innovations developed by early adopters.Trade islands,should they occur,would not be a permanent divide but rather a transitional phase towards a more integrated and dynamic global trade landscape enabled by AI.Nevertheless,the transition period could be sho

218、rtened or avoided altogether through coordinated action focused on enabling human-AI collaboration and facilitating human-human collaboration to reach AI convergence.Human-AI collaboration Successful AI integration into global trade involves collaboration with(rather than the replacement of)human ex

219、pertise.AI brings unprecedented capabilities in data analysis,pattern recognition and predictive modelling,while human judgement remains essential for innovation,contextual understanding,ethical oversight and strategic decision-making.Combining these skill sets is vastly more powerful than either en

220、tity operating in isolation.When implemented effectively,AI handles intensive data processing,freeing humans to focus on higher-order challenges that require both quantitative insight and qualitative understanding from strategic planning to ethical oversight of complex systems.AI convergence therefo

221、re depends on thoughtful workforce development by businesses and governments to train the current workforce while preparing future generations for an AI-enhanced landscape.Robust digital infrastructure and innovation incentives are key to human-AI collaboration,enabling humans to continuously improv

222、e the technologys capabilities.By investing in computational resources,data systems and innovation frameworks,environments where technology and human creativity can work together effectively can flourish.Strategic incentives,such as research funding,collaborative platforms and training programmes,en

223、courage diverse talents toengage with AI,ensuring that its development isrepresentative and remains responsive to human needs.This can transform AI from a standalone technology into a powerful collaborative tool that enhances human problem-solving and potential.Human-human collaborationCollaboration

224、 among humans,through international regulatory coordination and PPPs,is essential to reaching AI convergence.Regulatory coordination can establish common standards and frameworks to ensure interoperability between systems and establish trust in the technology.International bodies like the World Trad

225、e Organization(WTO)have a role to play in minimizing regulatory fragmentation and ultimately preventing AI divergence.Regulatory frameworks must be comprehensive,focusing on interoperability,security and transparency to mitigate fragmentation while remaining flexible enough to accommodate AI innovat

226、ion a difficult balancing act,given the speed of innovation.The implementation of AI across global trade at every scale is inevitable,given the gains the technology offers businesses and governments.Collaboration among humans,through international regulatory coordination and PPPs,is essential to rea

227、ching AI convergence.31Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTechEnsuring a comprehensive yet flexible regulatory framework requires an understanding of the various competitive incentives from all stakeholders in global trade,as well as an understanding of the

228、 technology itself.PPPs emerge as essential connective infrastructure to address unaligned incentives and a lack of information,bridging governmental oversight,private sector innovation,academic expertise and civil society concerns.PPPs like the TradeTech Global Initiative can facilitate AI converge

229、nce by:Connecting diverse stakeholders to alignincentives Informing the development of frameworks through regulatory sandboxes Raising awareness on key issues through thought leadership material Shedding light on cutting-edge developments through start-up accelerators Sharing best practices from exp

230、erts and regulators around the world Pushing for collaborative approaches toAIimplementation While preventing fragmentation in the face of ongoing global challenges is difficult,collaboration has never been easier,thanks to AI.AI empowers humans in previously unthinkable ways,enabling the synthesis

231、of vast quantities of information in real time and facilitating communication in any language on the planet.As these tools develop,they will inevitably continue to revolutionize trade.However,in navigating the complexities of AI implementation in global trade,success will be measured not by technolo

232、gical advances but by the ability to use AI to create a more inclusive,sustainable and efficient interconnected global trade network.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech32ContributorsAcknowledgementsWe would like to thank the team at the Ministry of Econ

233、omy and Yasar Jarrar,Racha Mourtada and Barkha Shah for their review,as well as the following contributors who provided input orfeedback:Miishe Addy Chief Executive Officer,Jetstream Africa,Ghana Paul BakerChief Executive Officer,International Economics ConsultingJohn BescecDirector,Customs and Trad

234、e Affairs,MicrosoftMariana Bock-LosadaChief Growth Officer,Digital Container Shipping AssociationLaetitia CailleteauManaging Director,Accenture,United KingdomNoura Al DhaheriChief Executive Officer,Maqta Technologies Group and Digital Cluster,AD Ports GroupThomas Fahey Associate Director,Supply Chai

235、n and Operations,AccentureXiuju FuSenior Principal Scientist,Institute of High Performance Computing,Agency for Science,Technology and Research(A*STAR)Emmanuelle GanneSenior Analyst,Economic Research and Statistics Division,World Trade Organization(WTO)Baghdad GherrasChief Digital Officer,Medad Hold

236、ing,United Arab EmiratesPuay Guan GohAssociate Professor and Academic Director,MSc Smart Industries and Digital Transformation,NUS Business SchoolRebecca HardingStrategic Trade Advisor and Analyst,RebeccanomicsPeng JiangAssociate Professor,Department of Industrial Engineering and Management,Sichuan

237、UniversitySebastian KlotzManager,Trade,Sustainability,and Technology,PricewaterhouseCoopers Paul KlimosGeneral Counsel,ProvableJeppe KobberDirector and Head of Implementation Customs AI,DSVLead authorsAlexandre MarchandSpecialist,TradeTech,World Economic ForumJimena SoteloLead,TradeTech,World Econom

238、ic ForumWorld Economic Forum Mario CanalesSpecialist,TradeTechSean DohertyHead,International Trade and InvestmentNitin SharmaLead,Digital Trade and Venture,TradeTechTim StekkingerHead,TradeTechWe would like to thank colleagues in the Centre for the Fourth Industrial Revolution,namely Karla Yee Ameza

239、ga,Maria Basso and Benjamin Larsen for their review.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech33Wolfgang LehmacherOperating Partner,Industrial Innovation Partners,Anchor GroupMikael Lind Senior Strategic Research Advisor,Safety,RISESaif Al Mehrzi International

240、 Relations Executive,Ministry of Cabinet Affairs of the United Arab EmiratesAyoub NaciriGBS Blockchain Practice,IBMSandra NolascoFounder and Chief Executive Officer,Twinco CapitalMuhammad Irfan SarfrazChief Knowledge Officer,Agility James RilettChief Product Officer,Open MineralYousef Al RiyamiChief

241、 Digital Officer,Trade and Services,Maqta Technologies GroupMara SaavedraChief,Staff,Key State CapitalSarfaraz Sait IsmailHead,Marcoms,Haifin,United Arab EmiratesThibault SerletDirector,Research,Key State CapitalJonpaul Simpson Director,Trade Facilitation,Abu Dhabi Department of Economic Development

242、Riyan SherifPublic Affairs and Global Engagement Manager,DP WorldPeter SwartzCo-Founder and Chief Science Officer,Altana AILee TaroneFounder and Chief Executive Officer,Voy Finance Quan ZhaoSenior Programme Officer,Trade Policy,International Trade CentreProcedures and Facilitation Sub-DirectorateWor

243、ld Customs Organization(WCO)ProductionMary BridgesFreelance WriterLaurence DenmarkCreative Director,Studio MikoJay KellyDesigner,Studio MikoWill LileyEditor,Studio MikoArtificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech34Endnotes1.International Data Corporation(IDC).(2

244、024).IDC:Artificial Intelligence Will Contribute$19.9 Trillion to the Global Economy through 2030 and Drive 3.5%of Global GDP in 2030.https:/ impact of artificial intelligence on international trade.Brookings.https:/www.brookings.edu/articles/the-impact-of-artificial-intelligence-on-international-tr

245、ade/.3.World Trade Organization(WTO).(2024).Trading with intelligence:How AI shapes and is shaped by international trade.https:/www.wto.org/english/res_e/publications_e/trading_with_intelligence_e.htm.4.Miroudot,S.and C.Cadestin.(2017).Services In Global Value Chains:From Inputs to Value-Creating Ac

246、tivities.Organisation for Economic Co-operation and Development(OECD).https:/www.oecd-ilibrary.org/trade/services-in-global-value-chains_465f0d8b-en.5.Dubai Multi Commodities Centre(DMCC).(2024).The Future of Trade.https:/ technology to get on top of Scope 3 emissions.https:/ AI vs.predictive AI:Wha

247、ts the difference?IBM.https:/ EV Battery Supply Chain Explained.Rocky Mountain Institute(RMI).https:/rmi.org/the-ev-battery-supply-chain-explained/.9.Alicke,K.,V.Dilda,S.Grner,L.Mori,et al.(2021).Succeeding in the AI supply-chain revolution.McKinsey&Company.https:/ Economic Forum.(2023).Shared intel

248、ligence for resilient supply systems.https:/www.weforum.org/publications/shared-intelligence-for-resilient-supply-systems/.11.Organisation for Economic Co-operation and Development(OECD)and European Union Intellectual Property Office(EUIPO).(2021).Global Trade in Fakes:A Worrying Threat.https:/euipo

249、.europa.eu/tunnel-web/secure/webdav/guest/document_library/observatory/documents/reports/2021_EUIPO_OECD_Report_Fakes/2021_EUIPO_OECD_Trate_Fakes_Study_FullR_en.pdf.12.Mehta,D.(2024).Amazons latest Brand Protection Report:How were cracking down on counterfeit products.Amazon.https:/ Steps Up Fight A

250、gainst Hidden Links and IPR Vigilance Post Reorganization.Alibaba.https:/ of the Fake Annual Report 2024.https:/ EVs tell us about global trades troubles.Hinrich Foundation.https:/ al.(2023).ADB Briefs No.256:2023 Trade Finance Gaps,Growth,and Jobs Survey.Asian Development Bank(ADB).https:/www.adb.o

251、rg/sites/default/files/publication/906596/adb-brief-256-2023-trade-finance-gaps-growth-jobs-survey.pdf.17.Ibid.18.Ibid.19.U.S.Customs and Border Protection.(n.d.).Uyghur Forced Labor Prevention Act(UFLPA).https:/www.cbp.gov/trade/forced-labor/UFLPA.20.Altana.(2023).United States Customs and Border P

252、rotection Chooses Altana to Help Drive Implementation of the Uyghur Forced Labor Prevention Act.https:/altana.ai/resources/united-states-customs-and-border-protection-chooses-altana.21.Klotz,S.,S.Barr and J.Sotelo.(2024).Digitalization is disrupting global trade heres how AI can help customs and bus

253、inesses to respond.World Economic Forum.https:/www.weforum.org/stories/2024/06/ai-is-already-helping-businesses-to-buy-and-sell-internationally-heres-how-technology-can-disrupt-global-trade-even-more/.22.World Trade Organization(WTO).(2024).Trading with intelligence:How AI shapes and is shaped by in

254、ternational trade.https:/www.wto.org/english/res_e/publications_e/trading_with_intelligence_e.htm.23.Bernhofen,D.M.,Z.El-Sahli and R.Kneller.(2013).Estimating the Effects of the Container Revolution on World Trade.The Ludwig-Maximilian University of Munichs Center for Economic Studies and the Ifo In

255、stitute.https:/www.cesifo.org/DocDL/cesifo1_wp4136.pdf.Artificial Intelligence for Efficiency,Sustainability and Inclusivity in TradeTech35World Economic Forum9193 route de la CapiteCH-1223 Cologny/GenevaSwitzerland Tel.:+41(0)22 869 1212Fax:+41(0)22 786 2744contactweforum.orgwww.weforum.orgThe World Economic Forum,committed to improving the state of the world,is the International Organization for Public-Private Cooperation.The Forum engages the foremost political,business and other leaders of society to shape global,regional and industry agendas.