AI has rapidly emerged as a widely used technology in law and business, including in the supply chain industry. Its ability to quickly process and analyze vast amounts of diverse data (referred to as big data) has made AI a valuable resource for companies and transformed how they engage with their suppliers. At a time when supply chains continue to be volatile and complex, AI is increasingly being integrated by supply chain participants to:
This article examines the most common use cases of AI in the supply chain, giving examples of how companies and providers are integrating this diverse and rapidly improving technology into their operations (for more on the supply chain and its participants, see Supply Chain: Overview on Practical Law). This article also analyzes how attorneys supporting the supply chain function can use GenAI and other solutions in their practice.
(For a collection of resources to assist counsel on AI and emerging legal issues and regulations across various practice areas, see AI Toolkit (US) on Practical Law; for more on key AI and machine learning concepts, see AI and Machine Learning: Overview on Practical Law.)
Use of AI in the Supply Chain
Companies continue to integrate AI technologies into all facets of the supply chain. The emergence of agentic AI systems, which are capable of operating more autonomously than traditional AI solutions, will continue to drive efficiencies and enable more sophisticated automation in supply chain management.
This article reviews how companies are using AI in demand forecasting, procurement and supplier risk management, manufacturing, inventory management and warehousing, transportation and logistics, customs clearance and compliance, customer service and support, fraud detection and prevention, sustainability, and legal support. While these are some of the most common use cases, this is not an exhaustive list, and some overlap in these areas is unavoidable (for example, use of AI in demand forecasting directly affects production planning and inventory management).
Demand Forecasting
Companies are using machine learning and predictive analytics to forecast demand for their products. With AI's ability to process vast amounts of data, companies can expand their forecasting from the more traditional, manual, and short-term methods to AI-driven demand forecasting that can produce short- and long-term forecasts with better accuracy.
AI algorithms can predict future demand with greater accuracy by analyzing:
This enables businesses to manage inventory levels more efficiently, reducing both overstock and stockouts. For instance, retailers like Walmart and Amazon use AI-driven demand forecasting to anticipate customer needs and adjust inventory accordingly. By leveraging machine learning models, they can predict which products will be in high demand during specific periods, such as holiday seasons or promotional events.
AI-driven demand forecasting also benefits manufacturers. For example, Unilever and Pfizer use AI to predict demand for their products, enabling them to:
Agentic AI systems are able to take demanding forecasting one step further by autonomously adjusting production schedules without the need for approvals by a human.
(For a model clause covering forecasts, with explanatory notes and drafting tips, see General Contract Clauses: Forecasts on Practical Law.)
Procurement and Supplier Risk Management
AI can help businesses assess and manage supply chain risk based on a variety of factors, including:
For example, SAP Ariba uses machine learning and natural language processing to provide real-time information on supplier operations. This enables their customers to:
Quick and effective vendor selection and due diligence can be particularly important in a crisis. AI technology can help manage the risk of supply chain disruptions and ensure that raw materials and components are continually available. For example, Unilever and Siemens use AI software that generates a list of potential new suppliers by sifting through numerous categories of data, such as:
In addition to helping locate alternate reputable sources of supply, AI can help evaluate if existing suppliers can provide additional raw materials or components. For example, Koch Industries uses AI to analyze data from transactions with current suppliers to determine if they could also supply other components. This reduces the costs associated with seeking new suppliers, such as the need for requests for proposals (RFPs) and new supplier onboarding.
AI can be used to manage supply chain-related risks more generally. For example, Resilinc uses AI-driven event monitoring to analyze large amounts of news data that may affect the supply chain, such as weather events or political upheaval. This allows companies to better deal with and plan for potential disruptions, providing for more resilient supply chains.
(For more on managing supply chain risk, see Managing Supply Chain Disruptions in a Crisis and Managing Supply Chain Disruptions in a Crisis Checklist on Practical Law.)
Manufacturing
AI allows companies to accelerate their manufacturing processes from design to commercialization. Companies can train GenAI models on their own data to identify ways to improve efficiency and productivity. Some companies, such as Siemens, GE, and Bosch, integrate advanced technologies like big data analytics and the internet of things (IoT) into highly efficient and flexible smart manufacturing processes. They optimize operations, improve quality, and reduce waste through:
AI can be used to manage inventory more efficiently. AI algorithms use historical sales data, market trends, and real-time inventory levels to predict when and how much stock should be replenished. Businesses can use AI to:
For example, companies like Gather AI deploy drones through warehouses to photograph inventory. These systems read barcodes, text, and other information and automatically compare this data with data in the warehouse management system. This process allows warehouse operations managers to see inventory data in real time, without having warehouse associates do these tasks manually.
AI can also be used to automate warehouse operations. AI systems like robotics and machine learning optimize storage space in warehouses, minimize errors, and reduce labor costs. For example:
Companies can also use deep-learning AI systems to make suggestions for alternatives when an ordered product is out of stock. For example, to recommend a substitute product for its online grocery orders, Walmart uses an AI system that considers numerous variables in real time, such as:
The system also provides the location of the substitute items in the store, leading to improved worker efficiency and higher rates of customer acceptance of substitutions.
(For more on the role of inventory management in the supply chain, see Inventory and Inventory Management on Practical Law; for more on the role of warehousing in the supply chain, see Warehousing on Practical Law.)
Transportation and Logistics
AI technologies like predictive analytics are making logistics and transportation more efficient. AI-powered systems can synthesize and analyze large amounts of historical data and real-time information. Algorithms can analyze traffic patterns, delivery schedules, and weather conditions to determine the most efficient routes.
This helps improve delivery times and minimize delays, allowing companies to reduce transportation costs. For example, Maersk uses AI to analyze data from its ships, ports, and warehouses to optimize the flow of goods through its network. Their AI system predicts when a ship might be delayed and reroutes it to avoid high traffic times.
Improved efficiency in deliveries also has the benefit of reducing fuel consumption and having a softer impact on the environment. UPS uses an AI platform for route optimization, which has resulted in substantial savings in fuel costs. Similarly, DHL uses AI to plan dynamic and optimized routes, allowing for improved balancing of freight across their routes and resulting in fewer detours or delays.
AI can also help with freight matching and load optimization. Companies like Trimble use their AI-driven transportation management software to pair shippers with carriers that have available space. Their AI-driven platform increases efficiency and helps reduce waste by limiting empty miles (miles that trucks travel without cargo, which can amount to 40 percent of total miles).
Autonomous trucks continue to be tested for transportation and deliveries. They are not yet being mass-produced because providers struggle with funding and ensuring that these vehicles are cost effective and safe. However, some startups have been successful in raising money and are already using autonomous trucks in providing transportation and delivery services in limited markets. For example, Gatik has made hundreds of thousands of autonomous deliveries to Fortune 500 companies with a perfect safety record.
(For a collection of resources to assist counsel with legal issues related to transportation and logistics, see Transportation and Logistics Toolkit on Practical Law.)
Customs Clearance and Compliance
AI is increasingly being used to streamline the customs clearance process and ensure compliance with customs regulations. AI-powered systems help in automating document processing, classifying goods, and predicting duties and taxes. This helps reduce errors and expedites the clearance process. For example:
US Customs and Border Protection (CBP) officials also use AI to help screen cargo and assess risk, allowing them to inspect high-risk goods more thoroughly and clear lower-risk goods faster. AI models automatically identify objects in video and images and send alerts in real time when they detect an anomaly. This helps CBP to stop illegal goods from entering the country.
(For more on key importing requirements and complying with them, see Importing Goods: Overview (US) and Core Elements of an Import Compliance Program on Practical Law.)
Customer Service and Support
Businesses are using AI-powered chatbots and virtual assistants to handle common customer inquiries, such as order status, delivery times, and product availability. This is happening in both B2B (business-to-business) and B2C (business-to-consumer) environments. For example:
These systems can provide responses faster than human customer service representatives. They can improve customer satisfaction and allow employees to focus on more unusual or complex inquiries.
Fraud Detection and Prevention
Companies like IBM use machine learning algorithms to analyze large amounts of transactional data to detect unusual patterns and anomalies. This helps to identify fraudulent activities within the supply chain and reduce losses. Similarly, transportation companies use data analytics and machine learning to help pinpoint outliers in their operations. They can use AI to analyze vast amounts of operational data to detect uncommon activities or patterns that may indicate theft, fraud, or scams.
Businesses are also able to leverage AI with blockchain technology for more secure transactions in the supply chain. For example, IBM Blockchain helps minimize fraud by providing a transparent ledger that cannot be modified, where all transactions are recorded and accessible to authorized parties. This allows participants to:
Using AI technologies like predictive intelligence and data analytics, companies can assess and rate the sustainability practices of businesses. For example, EcoVadis's AI-driven platform enables companies to evaluate their own sustainability program and that of their supply chain partners. This allows companies to meet their environmental, social, and governance (ESG) goals and manage related risks by adjusting their own practices and engaging with their current suppliers or identifying new ones with higher ratings.
Similarly, SupplyShift uses AI to collect and analyze data from many sources to evaluate vendors' carbon emissions, labor practices, community impact, and efficiency in using resources like raw materials, water, and electricity. The company's algorithms and data analytics offer their clients transparency into their supply chains and facilitate making informed decisions about suppliers.
SAP Ariba's AI-driven procurement and risk management solutions are also useful in evaluating the sustainability-related performance of vendors (see Procurement and Supplier Risk Management above).
Counsel advising the supply chain function can also integrate AI into their practice. Attorneys can use AI to conduct research and produce initial drafts of legal documents. While legal professionals must exercise caution and verify answers provided by AI, they can leverage the technology to more efficiently complete low-risk or repetitive tasks and focus on tasks with higher added value, such as advising their clients on issues where business and legal judgment are needed. Increased efficiency also facilitates more optimal interactions between in-house and outside counsel. As AI solutions for legal services rapidly improve, the legal industry will continue to move from experimentation to increased adoption.
Automating Document Review
Counsel receive a variety of documents for review, ranging from emails to responses to RFPs to company policies. AI systems can help automate or expedite the document review process. For example, they can:
This helps counsel to:
For example, Thomson Reuters's HighQ combines advanced AI and data analytics to help with document review by extracting key information, identifying relevant clauses, and flagging potential risks or compliance issues. HighQ also automates repetitive tasks, such as document classification, tagging, and data extraction, to speed up document review and minimize errors. Similarly, Westlaw Edge's Quick Check allows users to upload a brief and have the AI analyze the content to identify relevant cases and statutes that might have been missed.
Counsel can leverage AI to support the supply chain function's contracting activities. When contracting with suppliers or other parties supporting the supply chain, counsel can use AI to help with various tasks throughout the lifecycle of a contract, including:
Company and Vendor Compliance
Counsel can use AI to review and analyze large volumes of data to help meet the compliance requirements of the company and its supply chain partners. Legal departments can use AI to:
For example, Thomson Reuters Regulatory Intelligence uses:
Users can create customized alerts and receive bespoke updates to stay abreast of developments in a specific area.
(For more on providing advice to the supply chain function on compliance matters, see Supply Chain: Overview on Practical Law; for more on leveraging AI solutions, see Using AI in Law Departments on Practical Law.)
Legal Research
AI-assisted legal research solutions can synthetize large amounts of legal materials to quickly surface relevant:
A key concern for attorneys using large language models (LLMs) to conduct legal research is accuracy. To address this concern and limit an LLM's hallucinations, providers have developed retrieval augmented generation (RAG). RAG is an AI framework that grounds the model in a particular database containing trusted, verified information. Using RAG with an LLM allows for:
For example, Search & Summarize PL uses RAG to only search the Practical Law website instead of the internet. The system provides a summarized answer based on Practical Law materials only. It also provides citations to the underlying resources, so that users can validate the system's response.
Search & Summarize PL is also specifically prompted to not make up answers. For example, when asked a question about an area of the law that is not covered by Practical Law, Search & Summarize PL is designed to respond by telling the user that it is unable to answer the question using the source materials.
Despite these measures to improve accuracy, no AI system is infallible. Attorneys should therefore:
AI-powered assistants for the legal industry are making an impact on legal practice by integrating skills such as legal research, analysis, and document summarization and review into one tool. For example, Thomson Reuters' CoCounsel 2.0 is an advanced AI-powered legal assistant that leverages LLM technology and Thomson Reuters' databases like Westlaw and Practical Law. Attorneys are able to use CoCounsel 2.0 with their company's databases and their own precedents as well.
CoCounsel 2.0 can:
AI Assistants, like all AI tools, can greatly increase attorney efficiency. However, attorneys should use them to supplement, rather than replace, their own expertise and judgment, and always verify their outputs for accuracy.
