Key Legal Insights from ąó´Ç±ô±đ˛â’s Automotive Team

Analysis by Julie Dautermann, Competitive Intelligence Analyst

Â鶹´«Ă˝ is here to help you through all aspects of rethinking your long-term business strategies, investments, partnerships, and technology. Contact the authors, your Â鶹´«Ă˝ relationship partner, or our Automotive Team to discuss and learn more. 

KEY DEVELOPMENTS

• Â鶹´«Ă˝ & Lardner announced the launch of its 2026 AI in Manufacturing & Supply Chain Series, a new initiative to help industry participants identify and manage the legal risks and business strategies arising from the profound shifts and innovations reshaping manufacturing and supply chain operations. The most recent article in the series is Legal Considerations for IP in Smart Manufacturing: Data Ownership, Trade Secret Risks, and Patenting AI-Assisted Inventions. Subscribe to get updates about new articles in this series.

• Â鶹´«Ă˝ & Lardner partners Nicholas Ellis and Vanessa Miller were featured in Supply & Demand Chain Executive for their article, .
• U.S. new light-vehicle sales are projected to reach a SAAR of 16.3 million units in May, representing an increase of 5.8% year-over-year, according to a from JD Power and GlobalData. The analysis notes that year-over-year comparisons remain challenging, as May 2025 results were lowered by accelerated purchases in the two preceding months in response to U.S. import tariffs. Total global sales in 2026 are forecast to decline 1.1% YOY to 91.1 million units, due to market challenges that include the ongoing conflict in the Middle East, and weakened sales within China.
• Plante Moran’s 2026 North American Automotive OEM-SupplierĚýWorking Relations Index® (WRI®) Study overall improvements across all six North American OEMs for the first time in the study’s history. Toyota and Honda are in the top two positions, followed by GM, Nissan, Ford, and Stellantis.

• Â鶹´«Ă˝ & Lardner provided an update for multinationals pertaining to new rules for Section 232 tariffs on steel, aluminum, and copper derivatives.
• The Justice Department filed on May 29 announcing a Court of International Trade order that granted all U.S. importers the ability to seek refunds on IEEPA tariffs that were deemed unlawful by the U.S. Supreme Court.

• The U.S. Court of International Trade will not stay its ruling prohibiting the collection of Section 122 10% global tariffs for certain plaintiff importers while the judgment is appealed, according to a report in . This follows a federal appeals court’s on a decision that had found global tariffs imposed by President Trump under Section 122 of the Trade Act were unlawful.

• The Trump administration is to be considering a proposal that would require 50% of the components and materials in an automobile to be produced by U.S. sources in order to qualify for preferential treatment under the United States-Mexico-Canada Agreement. The agreement currently requires 75% of a vehicle’s materials to come from North American sources to avoid certain tariffs and qualify for USMCA preferential treatment.Ěý

• A May 2026 from the National Association of Manufacturers (NAM) called for the swift renewal of the USMCA, while recommendingĚý process improvements in areas that include protecting cross-border investments, securing critical mineral manufacturing inputs, and strengthening certain USMCA provisions to “discipline the behaviors of state-owned enterprises.”

• Over a dozen Democratic U.S. senators the U.S. Trade Representative to prioritize concerns during the USMCA review that include preventing the offshoring of U.S. manufacturing jobs to Mexico, strengthening the enforcement of labor laws in Mexico, and addressing increased Chinese investment in Mexico and Chinese expansion in North American supply chains.

• Automotive analysts how USMCA negotiations may be affected by trade agreements that established certain levels of U.S. import tariffs on vehicles and parts from the European Union, Japan, and South Korea.
• The European Union of its trade deal with the U.S., clearing a significant hurdle ahead of President Trump’s threatened July 4, 2026 deadline to raise tariffs on European automobiles if the agreement is not finalized. The European Parliament and EU countries are expected to vote to ratify the finished text later this month.

• U.S. automakers and dealerships may encounter higher prices and shortages of synthetic motor oils due to resulting from the Iran war, according to the Independent Lubricant Manufacturers Association.

---

OEMs/SUPPLIERS  

• A UAW strike at a supplier plant in Michigan could impact pickup truck production for GM, according to a report in .
• Volvo from the U.S. Office of Information and Communications Technology and Services to import and sell cars with connected vehicle technology. A Commerce Department that went into effect on March 17, 2025 of connected vehicles and related hardware and software linked to China with a phased approach beginning with the 2027 model year. Volvo has several production plants in China and it is majority-owned by Zhejiang Geely Holding Group.

• The U.S. auto industry is at risk of becoming less competitive in global markets due to certain that emphasize the domestic market.
• summarized commentary from a number of auto suppliers in support of USMCA renewal.

• GM to move production of the Chevrolet Groove and Aveo away from China joint venture plants to its Ramos Arizpe production site in Mexico beginning in 2027. GM employs more than 23,000 workers across four manufacturing complexes in Mexico.

• ±·´Ç±ą±đ±ôľ±˛ő’ largest aluminum plant is on track to in June, following two fires last fall that have disrupted supplies for certain automakers.
• Stellantis a $70 billion five-year strategic plan that will include the launch of by 2030, as well as in the North American market priced under $40,000.

• One million prospective buyers have avoided purchasing a new vehicle since the start of the decade due to persistently high prices, according to a report in . Automakers are expected to continue pursuing higher profits by selling fewer, more expensive trucks and SUVs.

• A number of European automakers are to be considering options for sharing unused plant capacity with Chinese automakers to reduce costs. Chinese automakers a combined record 10% market share in Europe in April.
• Stellantis intends to of AI-powered digital twins at certain plants in North America this year.

---

AUTONOMOUS TECHNOLOGIES AND VEHICLE SOFTWARE

• Waymo a new autonomous vehicle purpose-built specifically for robotaxi use in partnership with ZhejiangĚýGeely Holding subsidiary Zeekr. Public rides will begin in parts of San Francisco, Los Angeles, and Phoenix.
• A recent from Accenture and Germany’s Center of Automotive Management assessed the past and planned launches of software-defined vehicles (SDVs) by select OEMs including Tesla, Toyota, BMW, Mercedes-Benz, Volkswagen, BYD, Geely, and Xpeng.
• Tesla of its full self-driving system – or FSD Supervised – in several countries that include China.
• BYD a 4 nanometer autonomous driving chip at its technology day in China. The automaker plans to invest over $14 billion in intelligent technology R&D over the next three years.
• Mercedes to launch its MB.Drive Assist Pro assisted driving system incertain German cities later this year. This will follow the availability of a limited version on certain CLA models in the U.S., and the 2025 debut on Mercedes models in China. The – which requires driver supervision at all times – was as a more cost-effective replacement to the automaker’s SAE Level 3ĚýDrive Pilot autonomous technology.
• in atypical traffic situations remains an area of focus for a number of robotaxi developers.

---

MARKET TRENDS AND REGULATORY

• Certain U.S. refineries are to be prioritizing the production of jet fuel and reducing gasoline output in response to global supply constraints caused by ongoing in the Strait of Hormuz. This may contribute to higher gasoline prices during the peak summer travel period.
• Used-car retailer Carvana is into new-car sales by acquiring certain Stellantis dealerships and selling vehicles online.
• A number of third-party vehicle marketplace sites have revised pricing displays in response to greater from the Federal Trade Commission on dealership advertising practices.
• The , a bipartisan five-year surface transportation reauthorization, includes a to impose annual registration fees on battery electric and plug-in hybrid electric vehicles to support highway funding. The bill introduced by U.S. House lawmakers on May 19 also theĚý Department of Transportation to establish performance-based safety standards for certain automated driving system-equipped commercial motor vehicles operating in interstate commerce.
• The Environmental Protection Agency delaying stricter emissions standards for light- and medium-duty vehicles until model year 2029.

---

HYBRID AND Electric Vehicles

• Toyota of a new EV for its Lexus brand due to a broader review of new projects and weak consumer demand.
• Hybrids are a fast-growing vehicle segment in the U.S., having in the last three years to 20% of total U.S. new-vehicle sales.
• Retail sales of passenger new energy vehicles (NEVs) in China year-to-date through mid-May are . The decline was to an increasingly saturated market, reduced subsidies, and weak consumer spending.
• In the , hybrid-electric models were the top choice among European Union consumers, accounting for 38.2% of the total new-vehicle registrations, followed by petrol and diesel cars (30.2%), battery-electric vehicles (19.7%), and plug-in hybrid electric vehicles (9.6%). Chinese brands for 15% of BEV sales in Europe in April.

More than five years since the start of the COVID-19 pandemic, manufacturers across the globe continue to face pressure and cost increases from volatile supply chains, including increasing cost for raw materials, labor, energy, and other inputs, as well as disruptions from natural disasters, wars, tariffs, export restrictions, and other governmental actions. These issues can be particularly acute for automotive suppliers in an industry that historically operates on a just-in-time basis while relying on long-term sole-source contracts that require a lengthy validation process and OEM customer approvals to switch suppliers. While there is no silver bullet, this article will consider some of the contracting and pricing strategies that companies can employ to mitigate their risk.

At the basic level, nearly every provision of a long-term contract allocates some share of risk to each side of the deal, and pricing is no different.

If the parties enter into a contract with a fixed price for the entire duration of the program, the seller is assuming the risk that the market price will go up (including increases due to a rise in cost to manufacture), while the buyer is assuming the risk that the market price will go down (including decrease due to reductions in the cost to manufacture). Any deviation from a fixed price represents different allocation of risk that may, or may not, be offset by changing the risk allocation provided for in other terms.

For example, a contract that allows the seller freedom to raise prices relieves the seller from the risk that its costs may increase substantially without an ability to increase prices accordingly, but it creates a new risk for the buyer that the seller may seek to increase the price beyond what is justified by the market, particularly if the buyer is obligated to purchase only from the seller. However, these risks can be offset by changing other clauses including, for example, giving the buyer a right to terminate the contract for convenience and go elsewhere if it can get a better price on the open market.

In a world where costs to manufacture are only increasing with few, if any, reductions, a fixed-price contract would seem, on its face, to impose significantly more risk on the seller. In many cases, this will be true. However, buyers also can suffer costs from this approach. Namely, sellers faced with a likelihood that their costs will go up may be left with little choice but to try and project what their cost structure will look like over the life of the contract and inflate their quotations, increasing the price that the buyer has to pay beyond what it might have paid if it was willing to share in some of the future risk.

The following list, while not exclusive, summarizes some of the more common alternative pricing clauses and strategies that many suppliers are now seeking to incorporate into their contracts to better approve the risks between buyers and sellers.

•  Indexing Clauses â€“ On the simpler end of the scale, many suppliers seek to negotiate indexing provisions in their agreements. Under an indexing provision, the price for the goods is adjusted (up or down) at regular intervals based on the changes in certain third-party indexes. Traditionally, such clauses have most often been used for products where raw material costs, such as steel, aluminum, or resins, make up a high proportion of the cost for the goods.  However, with costs for labor, energy, and other inputs becoming a more significant issue for many suppliers, parties can also consider more creative arrangements in which cost inputs beyond material, or even the full purchase price of the goods, may be adjusted based on measures of inflation or other indexes.
• Volume Bands/Commitments â€“ The parties may agree to pricing that is tied to anticipated purchase volumes, with the quoted price applying only if the buyer meets specified volume thresholds, forecasts, or minimum purchase commitments. If actual volumes fall below agreed bands or targets, the price may adjust upward to reflect the seller’s higher per-unit costs, lower economies of scale, or unrecovered capital and labor assumptions that were built into the original quotation. Conversely, higher volume bands may support lower pricing. These provisions can help align pricing with the commercial assumptions underlying the deal, but they should be drafted carefully to address how volumes are measured, the applicable review period, whether shortfalls caused by force majeure or buyer-directed schedule changes are excluded, and whether the adjustment is automatic or subject to further negotiation.
• Material Resale â€“ In the automotive industry, many OEMs utilize a form of material resale program under which the OEM uses its bargaining power and overall volumes to negotiate pricing with suppliers for steel, aluminum, and other raw materials which they then make available to their suppliers.
• Review, Adjustment and Negotiations â€“ In some cases, where parties are unable to agree on a specific formula or mechanism for future price adjustments, they may compromise on a more general statement that prices will be “reviewed” or that the parties will “negotiate” or “adjust” prices in the future. While such general clauses provide some level of protection for sellers against being locked into a fixed-price contract, they are often ambiguous and can lead to disputes down the road as far as what adjustments are actually required to be made.
• Multi-Sourcing â€“ While traditionally not common in the automotive industry, some suppliers are adjusting their purchasing strategy to include more multi-source arrangements, particularly for raw materials or other fungible inputs. Multi-sourcing can mitigate the risk of supply interruptions, while also providing suppliers with greater leverage over their sub-suppliers in any pricing discussions. However, multi-souring can also come with trade-offs, including potentially higher prices based on lower volume purchases, and challenges with ensuring that both contracts have sufficiently enforceable quantity terms that meet the requirements of the Uniform Commercial Code.
• Tariff Clauses â€“ Over the last 16 months, clauses seeking to address the impact of changing tariffs have become a particular point of emphasis for many suppliers. In their simplest form, such clauses may allow for a direct pass-through of tariff’s paid by the seller, or changing the location of delivery so that the buyer must assume responsibility to pay tariffs directly as the importer of record. However, accounting for tariffs paid on the end product itself often is only part of the picture, as sellers may face additional costs due to price increases imposed by sub-suppliers, which may require negotiation of additional cost shifting provisions.

Not all of these provisions are appropriate in all situations, and negotiation of such terms ultimately still depends on the relative leverage between the parties. However, these are examples that highlight the kind of creative contracting strategies that may be available to automotive suppliers in order to mitigate the risk posed by entering into long-term contracts in an environment that remains, and is expected to remain, volatile.

This article originally appeared on Ěýin May 2026.

Key Takeaways

• Adopt a hybrid IP protection strategy: Combine patents for core technical innovations with trade secrets for evolving know-how (like datasets and AI optimization methods), supported by clear contracts that define ownership and usage rights in collaborations.
• Strengthen trade secret governance in AI environments: Mitigate risks from “shadow AI” and data leakage by limiting use to approved tools, enforcing access controls, embedding strong contractual protections, and training employees on safeguarding sensitive information.
• Clarify data ownership and document AI-assisted inventions: Explicitly define data rights in agreements to avoid disputes and ensure patent eligibility by documenting human contributions in AI-assisted innovations, since only natural persons qualify as inventors.

The rapid deployment of artificial intelligence (AI) -driven manufacturing technologies presents a dual challenge for in-house counsel and business leaders: how to capture competitive advantage through innovation while managing the attendant intellectual property (IP) risks. As manufacturers integrate AI-driven robotics, Internet of Things (IoT)-enabled digital twins, and agentic AI systems into their operations, they are generating significant intellectual property assets and exposures that demand proactive legal strategy.

Three questions should keep every manufacturing executive up at night. First: Are we adequately protecting the innovations our engineering and design teams are developing with AI assistance — or are we leaving valuable IP exposed? Second: We’re generating massive volumes of operational data, but do we actually own it, and what can we legally do with it? Third: AI tools are proliferating across our workforce, often without oversight — how do we prevent a single employee prompt from exposing our most sensitive trade secrets?

This article, part of ąó´Ç±ô±đ˛â’s 2026 AI in Manufacturing & Supply Chain Series, distills practical legal frameworks at the intersection of AI innovation and IP protection, with strategies to secure IP rights, mitigate litigation risk, and navigate the contractual and regulatory complexities inherent in smart manufacturing.

1. Intellectual Property Protection Strategies for Smart Manufacturing Technologies

Effective IP management in smart manufacturing requires a layered strategy deploying patents, trade secrets, and contract rights in coordinated fashion. Consider automotive manufacturing: today’s vehicles are software-defined platforms more akin to iPhones than to the cars of a generation ago, and automakers must increasingly think like technology companies rather than traditional manufacturers. This shift is emblematic of the broader smart manufacturing landscape, where patents are even more essential for protecting discrete technical innovations, such as novel AI integrations with industrial control systems, proprietary sensor-fusion algorithms, and advances in predictive maintenance. However, many competitive advantages in AI-driven manufacturing lie in iterative know-how and continuously evolving systems. In some cases, such competitive advantages may be better shielded as trade secrets.

Leading manufacturers increasingly employ a hybrid approach: leveraging patent protection to protect core inventions and create licensing opportunities, while leveraging trade secret protection measures to protect training methodologies, proprietary datasets, and optimization parameters. Copyright protection supplements this framework by covering software code and user interfaces, while trademarks safeguard branding around proprietary platforms.  However, implementing such a hybrid approach must involve broad awareness of innovations across the company, as well as proper precautions in place to maintain the secrecy of innovations earmarked for trade secret protection.

In multi-party relationships, IP allocation should be addressed explicitly at the outset of any joint development. Governing agreements should clearly delineate background IP (pre-existing assets each party brings to the relationship), foreground IP (jointly developed innovations), and rights to improvements. Field-of-use restrictions, licensing terms, and sublicensing limitations should be negotiated with precision to avoid downstream disputes.

IP due diligence must precede the scaling of any AI tool in manufacturing. With patent filings in AI-for-manufacturing surging globally, manufacturers that rely solely on vendor representations without freedom-to-operate analysis risk costly infringement exposure.

2. Trade Secret Safeguards and Risks in AI‑Driven Manufacturing Environments

The integration of AI into manufacturing operations has materially expanded the trade secret attack surface. Generative and agentic AI tools can rapidly codify tacit employee knowledge — process workarounds, optimization heuristics, institutional expertise — into digital form. The appeal of digitization is obvious: critical knowledge at your fingertips, instantly accessible across your organization. But every asset you can access instantly is an asset a threat actor can target. The more information you digitize, the larger your attack surface becomes. And the threat is not theoretical — breaches dominate every news cycle, with threat actors increasingly targeting IoT networks and AI training data.

“Shadow AI” has emerged as a : employees deploying unvetted generative AI tools that may transmit sensitive data to external servers or use proprietary inputs to train third-party models. The reality is that employees will use AI tools — the question is whether they use approved tools with appropriate safeguards or unsanctioned alternatives that expose the organization. Companies that fail to provide enterprise-approved AI solutions do not eliminate usage; they simply drive it underground, where the risks multiply. In manufacturing environments where production data is routinely shared across partners and platforms, the risk of inadvertent disclosure is acute.

Mitigating these risks requires a layered governance framework:

• Approved Tools: Restrict AI tool usage to enterprise-approved, internally hosted, or contractually controlled systems with appropriate confidentiality protections.
• Access Controls: Implement tiered access controls, data segmentation, and continuous monitoring of AI interactions to detect and limit unauthorized data flows.
• Contractual Protections: Require robust nondisclosure agreements with all vendors, employees, and contractors, coupled with exit protocols that mandate return or destruction of proprietary information.
• Training and Audits: Conduct regular trade secret audits and targeted employee training to ensure personnel understand what constitutes proprietary information — and the risks of disclosure through AI tools.
• Need-to-Know Policies: If information is not germane to the relationship, there is no reason to share it — regardless of the protections that may be available. The most effective trade secret programs are built on a principle of deliberate restraint: share only what is necessary, and no more.

Contracts with AI vendors should include confidentiality obligations, data-use restrictions, model training prohibitions, and audit rights. Where feasible, manufacturers should host models internally or utilize private cloud instances. These safeguards reduce leakage risk and strengthen potential trade secret misappropriation claims under the Defend Trade Secrets Act by demonstrating reasonable measures to maintain secrecy.

3. Data Ownership and Contractual Risk Allocation in Smart Manufacturing

Data generated by smart manufacturing systems — sensor outputs, digital twin telemetry, AI inference logs — represents both a strategic asset and a persistent source of commercial dispute. In multi-party relationships, ambiguity over data rights can quickly escalate: Who owns sensor data from a shared production line? Can one party use another’s data to train AI models? What restrictions govern data shared with cloud providers? Without clear contractual allocation, these questions invite costly litigation.

Agreements should address data ownership with specificity: classification of data as background (pre-existing) or foreground (generated during the relationship); allocation of rights to derivative works, including AI models trained on shared data; licensing scope and field-of-use restrictions; flow-down obligations for subcontractors; and indemnification for data-related IP or privacy claims.

Limitation-of-liability clauses should be calibrated to data-driven dispute risks, and dispute resolution provisions should reflect the parties’ commercial relationship. Manufacturers should also secure audit rights over data usage and negotiate deletion or anonymization requirements upon contract termination.

4. Patenting AI‑Assisted Inventions in Smart Manufacturing

The U.S. Patent and Trademark Office’s resolved considerable uncertainty for patent practitioners and innovators alike. Implementing the Trump Administration’s Executive Order on American AI leadership, the USPTO reaffirmed the longstanding statutory requirement: only natural persons may be named as inventors. AI systems are instruments of invention, not inventors. The determinative inquiry remains conception: a natural person must form “a definite and permanent idea of the complete and operative invention.”

For manufacturers deploying AI in process innovation, this guidance has immediate operational implications. Patent claims arising from AI-assisted work must be supported by clear evidence of human intellectual contribution, whether through prompt engineering, data curation, model validation, or the integration of AI outputs into a workable technical solution. Failure to document the human contribution invites rejection or exposes issued patents to invalidity challenges.

Manufacturers should implement protocols for contemporaneous documentation of AI-assisted invention: iteration logs, decision matrices reflecting human oversight, and records of modifications to AI outputs. This documentation supports patent prosecution and provides critical evidence in enforcement actions. Additionally, practitioners should monitor foreign patent regimes, several of which are developing divergent AI-inventorship standards that may affect priority claims and filing strategy.

Conclusion: IP as Strategic Infrastructure

As AI-driven manufacturing technologies mature, intellectual property must be treated not as a compliance exercise but as core strategic infrastructure. Manufacturers that implement clear pipelines for tracking innovations, deploy layered protection strategies tailored to their business objectives, fortify trade secret programs, and negotiate precise IP ownership terms will be positioned to capture the value of their innovations while minimizing litigation exposure and regulatory risk.

ąó´Ç±ô±đ˛â’s Manufacturing, Technology, and Intellectual Property attorneys regularly advise manufacturers on IP strategy, contract structuring, and dispute resolution in AI-enabled operations. For more information, please contact the authors or your Â鶹´«Ă˝ relationship partner.

Agentic AI refers to an artificial intelligence system designed to achieve specific goals with minimal human supervision. Unlike traditional AI models that operate within predefined constraints and require human intervention, agentic AI exhibits autonomy, goal-driven behavior, and adaptability.

Key Takeaways:

• Agentic AI systems capable of autonomous supply chain decisions are already being used by major companies like and Flexport, and adoption is accelerating across the manufacturing sector.
• Standard AI vendor contracts typically cap liability at fees paid and exclude consequential damages leaving manufacturers exposed when autonomous decisions trigger excess inventory, stockouts, unnecessary freight costs, or product damage.
• Manufacturers evaluating agentic AI should focus on system design (authority limits, override protocols, data quality controls) and contractual architecture (realistic liability caps, audit requirements, indemnification) to capture operational benefits while managing legal risk.

The Rise of the Autonomous Supply Chain Agent

For manufacturers evaluating agentic AI for supply chain operations, the technology landscape has shifted significantly. Where earlier AI tools provided recommendations for human evaluation, the newest agentic systems are designed to act autonomously—placing purchase orders, adjusting safety stock levels, selecting carriers, and re-routing shipments in real time, often minimal to no human approval. The adoption curve is steep: now uses agentic AI for autonomous inventory replenishment and shipment re-routing across its fulfillment network, while Flexport’s AI agents autonomously manage approximately 40% of its freight forwarding operations, including dynamic shipment optimization and exception handling during disruptions.

The operational benefits are substantial—faster response times, reduced labor costs, and the ability to react to disruptions at machine speed. But so are the risks. Unlike predictive tools that flag anomalies for human evaluation, agentic systems act on their own conclusions. When those conclusions are wrong because the underlying data was flawed, the model’s logic was miscalibrated, or the system lacked contextual judgment, the financial and legal consequences can be immediate and difficult to unwind.

When the Agent Gets It Wrong: Four Scenarios Manufacturers Should Anticipate

Manufacturers considering agentic AI should anticipate the types of failures these systems can produce. The following scenarios illustrate some common risk categories that should inform system design, oversight protocols, and contractual protections:

1. Excess Inventory. An agent misreads duplicated or erroneous demand data and autonomously places excess purchase orders, leaving the manufacturer with carrying costs, write-downs, and potential disputes over order cancellation.
2. Stockouts and Line Stoppages. An agent optimizing working capital reduces safety stock based on outdated supplier data, eliminating the buffer needed when demand materializes and triggering line stoppages, missed deliveries, and premium freight charges.
3. Unnecessary Freight Costs. A logistics agent misinterprets a data lag as an inbound delay and authorizes expedited air freight at significant cost, only for the original shipment to arrive on time.
4. Product Damage From Re-Routing. An agent re-routes a temperature-sensitive shipment to avoid a forecasted bottleneck, but the alternative route passes through extreme temperatures, resulting in damaged goods and potential product liability exposure.

Why Existing Contracts May Not Protect You

Manufacturers facing these scenarios will naturally look to their AI vendor agreements for recourse. Standard AI vendor contracts typically cap liability at fees paid, which are often just annual subscription costs. However, a single errant autonomous decision can trigger losses many times over. Consequential damages waivers may bar recovery for the very harms—excess inventory, expedited freight, downtime, lost production—that agentic systems are most likely to cause.

Causation challenges are compounded in the agentic context. When an autonomous agent decides and acts, the manufacturer must untangle whether the failure originated with the AI model’s logic, the data inputs, the system’s configuration, or the absence of oversight. This could implicate the vendor, IT team, data providers, and system integrator simultaneously. Technology may explain why a decision was made, but it will not excuse the consequences.

Designing Systems and Contracts to Manage Agentic AI Risk

Manufacturers considering or already deploying agentic supply chain AI should focus on both system design and contractual architecture to manage these risks before disputes arise.

• Define Autonomous Authority Limits. Manufacturers should establish clear thresholds with respect to dollar values, volume quantities, and routing changesthat the agent must escalate to human review. These limits should be documented and reflected in the vendor agreement.
• Build in Override and Kill-Switch Protocols. Agentic systems should include monitoring dashboards and manual override capabilities. Contracts should specify the vendor’s obligation to provide these and allocate responsibility for losses when overrides are unavailable.
• Address Data Quality at the Source. Many scenarios originate not with the AI model but with the data feeding it. Flawed inputs lead to flawed decisions. Contracts should allocate data quality responsibilities clearly by specifying which party owns validation and what happens when quality falls below agreed thresholds. For a deeper analysis on data quality issues, see our prior article on this topic.
• Negotiate Liability Structures That Reflect Autonomous Decision-Making. If the system acts without human approval, the vendor’s liability framework should reflect that risk, including realistic liability caps, carve-outs from consequential damages waivers, indemnification for third-party claims, and appropriate insurance requirements.
• Maintain Audit Trails and Decision Logs. The ability to reconstruct why the agent made a decision will be critical for both defending against claims and pursuing the vendor. Contracts should require accessible decision logs and vendor cooperation with audits.

Positioning Your Organization for the Agentic Future

Agentic AI in supply chain operations is already being deployed. Manufacturers that take a deliberate approach to system design, governance, and contractual risk allocation will be better positioned to capture the benefits while limiting litigation exposure. When an autonomous agent makes a costly mistake, the dispute will be resolved under contracts and established legal principles—not algorithms.

Â鶹´«Ă˝ & Lardner’s Manufacturing, Supply Chain, and Artificial Intelligence teams are available to help organizations navigate the evolving risks of agentic AI in supply chain operations.

Key Legal Insights from ąó´Ç±ô±đ˛â’s Automotive Team

Analysis by Julie Dautermann, Competitive Intelligence Analyst

Â鶹´«Ă˝ is here to help you through all aspects of rethinking your long-term business strategies, investments, partnerships, and technology. Contact the authors, your Â鶹´«Ă˝ relationship partner, or our Automotive Team to discuss and learn more. 

KEY DEVELOPMENTS

• Â鶹´«Ă˝ & Lardner announced the launch of its 2026 AI in Manufacturing & Supply Chain Series, a new initiative to help industry participants identify and manage the legal risks and business strategies arising from the profound shifts and innovations reshaping manufacturing and supply chain operations. The most recent article in the series is Digital Twin Technology and Predictive Analytics in Manufacturing Supply Chains: Preventing Data-Driven Supply Chain Disputes. Subscribe to get updates about new articles in this series.

• President Trump he will raise U.S. import tariffs on European cars and trucks to 25% from 15% if the 27-member bloc does not formally approve by July 4, 2026 a previously negotiated framework trade deal.

• As of May 11, the Trump administration was more than $35.5 billion in refunds to importers whose tariff-refund claims were approved after the U.S. Supreme Court ruled the president’s IEEPA-based tariffs were unlawful.

• A federal appeals court a decision that had found President Trump’s new global tariffs—imposed under Section 122 of the Trade Act—were unlawful.
• In a recent letter, seven major automotive trade associations the U.S. Trade Representative to extend the U.S.-Mexico-Canada Agreement (USMCA), rather than pursuing separate bilateral deals.

• Automotive News provided an on the Trump administration’s potential priorities for USMCA negotiations with Mexico.
• A number of key stakeholders in the Canadian auto industry USMCA negotiations to extend past the July 1, 2026 timeline for the agreement’s scheduled review.Ěý

• GlobalData its 2026 U.S. light-vehicle sales forecast to 16.0 million units, representing a downward adjustment of approximately 100k units. The stand-off between the U.S. and Iran was cited as the reason for the revision, and GlobalData stated, “While we continue to assume that a solution will eventually be found and that oil prices will fall later in the year, there is enough uncertainty to lead us to be more cautious with the 2026 forecast.”

• and warned that shortages of key fuels in certain markets, and escalating prices,are a growing risk in the coming weeks if shipping does not resume in the Strait of Hormuz.

---

OEMs/SUPPLIERS  

• More than 60 auto suppliers operating in the U.S. are owned by companies located in China, according to from consulting firm AlixPartners.
• For fiscal year ending March 31, 2026, Toyota an operating loss in North America for the first time in sixteen years and noted it was unable to fully offset the impact U.S. import tariffs. The automaker’s most recent guidance is lower profit due to higher costs of fuel, transportation and paint due to the Iran war, as well as the ongoing impact of U.S. import tariffs.

• Honda its first annual profit loss since becoming a publicly traded company approximately 70 years ago. The automaker’s $2.7 annual profit loss was due largely to its costly retreat from ambitious EV targets.

• GM to a $12.75 million settlement over allegations it illegally sold the location and driving behavior data of California drivers to third-party data brokers. This represents the largest California Consumer Privacy Act (CCPA) penalty to date.

• Toyota is to be considering a $2 billion investment to expand production in San Antonio, Texas.
• Stellantis a partnership with Leapmotor to jointly develop vehicles for the European market at two existing factories in Spain.

• German auto-parts maker Schaeffler to deploy “thousands” of humanoid robots from British AI and robotics company Humanoid in its factories beginning in late 2026.
• Canadian union Unifor is to begin negotiations over a new collective bargaining agreement with Ford on June 22, 2026.

---

AUTONOMOUS TECHNOLOGIES AND VEHICLE SOFTWARE

• A number of expressed reduced confidence in data monetization, citing regulatory hurdles, limited consumer demand, and weaker-than-expected returns.
• AI agents capable of autonomous decision-making in certain automakers’ vehicle infotainment systems as soon as later this year.
• GM is utilizing AI technologies to vehicle design and autonomous vehicle development.
• A filed this month in New Jersey federal court alleges defects in Subaru’s “EyeSight driver assist technology” may cause certain models to abruptly stop without reason and increase the risk of collisions.

---

MARKET TRENDS AND REGULATORY

• Morgan Stanley U.S. gasoline inventories may fall below 200 million barrels by August 2026, representing a record seasonal low.
• A from Brown University’s Watson Institute estimates that U.S. households have paid over $300 in additional costs per household related to higher gasoline and diesel prices following the start of the Iran war.
• First-quarter 2026 total light-vehicle sales within China to 4.9 million units, while exports rose 58% YOY to 2.1 million units. According to data from the China Passenger Car Association, over 50% of China’s were EVs or hybrids, and this represented the first time the nation’s new-energy vehicle exports surpassed those of internal combustion engine vehicles.Ěý

---

HYBRID AND ELECTRIC VEHICLES

• Tesla approximately $250 million in its Berlin-Brandenburg factory to increase battery cell production for EVs.
• Stellantis with Dongfeng to produce EVs in China under the Peugeot and Jeep brands. The companies will invest up to $1.2 billion in the initiative.
• At the , hybrid vehicles on average had the lowest days’ supply and were the quickest to sell from U.S. dealerships when compared to other types of powertrains.Ěý
• A new indicates that hybrid vehicles, like battery electric vehicles (BEVs), experience reduced fuel efficiency in cold weather. On average, hybrids lose 22.8% fuel economy in freezing temps, while EVs decline by 35.6% MPGe.
• reported on ąó´Ç°ů»ĺ’s efforts to develop an all-electric pickup truck that can compete with Chinese brands and will be priced at approximately $30,000.
• Michigan may be offering an incentive package to electric pickup truck startup Slate Auto to support expansion in the state, according to a report in .
• including BYD, Chery, Geely, and Zeekr are reported to be making progress toward the launch of EV sales in Canada .

A digital twin is a dynamic, virtual representation of a physical asset, process, or system that is continuously updated by real‑world data.

Key Takeaways:

• Manufacturers are increasingly implementing “digital twins” and AI predictive analytic technology into the supply chain. These technologies have great potential but also introduce new legal risks.
• Data quality is the foundation of digital twin reliability—flawed inputs lead to flawed decisions, and new technology will not excuse contractual breaches or legal harms.
• As AI-driven tools increasingly influence operational decisions, proactive contract review and governance can help manufacturers identify and mitigate risks before disputes arise.

By 2026, “digital twins” and AI‑driven predictive analytics have become essential tools in supply chain and industrial manufacturing. Specifically, the digital twin in manufacturing represents a paradigm shift in how organizations manage production planning, supplier selection, capacity forecasting, inventory optimization, and quality control across global supply chains. For example, General Motors announced in March 2025 that it was working with NVIDIA Omniverse to create digital twins of assembly lines, and in March 2026 Delta Electronics announced a similar plan to monitor factory HVAC, lighting, and energy savings. As adoption accelerates, organizations should understand both the strategic advantages and the legal issues these technologies present.

The promise is compelling: a near‑real‑time virtual replica of factories, tooling, logistics lanes, and even supplier operations that can predict disruptions before they occur and promote efficiency. However, the legal risks embedded in these systems deserve careful attention—particularly when the data feeding them is inaccurate, incomplete, biased, or contractually misaligned. Organizations that recognize these vulnerabilities early can take steps to mitigate exposure before disputes arise.

How Digital Twin Technology Is Transforming Manufacturing Supply Chains

Organizations deploying digital twins and predictive models should understand the range of applications now common in manufacturing supply chains:

• Factories, production lines, and tooling performance
• Supplier capacity, lead times, and financial or operational risk
• Inventory levels across multi‑tier supply networks
• Logistics routes, transportation costs, and congestion
• Quality metrics, process deviations, and failure modes

Predictive analytics layered on top allow manufacturers to simulate scenarios before acting—for example, reallocating volume away from a supplier flagged as “high risk,” delaying capital investment based on forecasted demand shifts, or adjusting production schedules in response to perceived downstream slowdowns. Organizations should recognize that these capabilities, while powerful, depend heavily on data quality.

Importantly, these outputs are no longer confined to planning meetings. They directly influence purchasing decisions, supplier awards and terminations, volume commitments, and compliance determinations. This operational integration means that data quality issues can have immediate legal and business consequences.

Data Quality Risks in Digital Twins and Predictive Analytics for Manufacturing

Digital twins and predictive models depend on vast amounts of data, but in real‑world manufacturing environments, that data is often imperfect. Organizations should assess their data sources for the following common vulnerabilities:

• Supplier‑provided data that is self‑reported, delayed, or selectively curated
• Sensor and IoT data affected by calibration errors or inconsistent measurement standards
• Historical datasets that fail to reflect post‑pandemic demand volatility, reshoring initiatives, or geopolitical constraints
• Third‑party market data embedded with assumptions misaligned to contractual obligations

Digital twins and predictive models are only as good as the data they rely upon. Organizations should be aware that flawed data can produce outputs that are directionally wrong—sometimes subtly, sometimes catastrophically. A manufacturer may underproduce based on inaccurate demand forecasts, overcommit to a single supplier based on inflated capacity reporting, or prematurely disengage from a counterparty flagged as risky by an opaque algorithm. Identifying these failure modes before they occur is essential to managing downstream risk.

From a legal perspective, the technology may explain why a decision was made, but it likely will not excuse the consequences if the decision breaches a contract or creates a legally cognizable harm. Organizations should consider how their use of AI-driven tools will be perceived in the context of potential litigation.

Legal and Contract Risks From Predictive Analytics in Supply Chain Decision-Making

Manufacturers and suppliers should be aware that disputes tied to AI‑assisted decision‑making often take familiar legal forms. Proactive identification of these risk areas can help organizations strengthen their contractual protections and operational safeguards.

Review Supply and Volume Agreement Exposure

Manufacturers should assess whether their supply and volume agreements account for data-driven adjustments. When predictive analytics drive reductions in orders or changes in sourcing, suppliers may look to contractual minimums and allocation clauses. Consider whether contracts explicitly address forecasting assumptions and permit data-driven adjustments—otherwise, volumes that drop below agreed levels may create liability.

Evaluate Termination and Default Procedures

Manufacturers that rely on predictive tools to monitor supplier performance and financial health should evaluate the accuracy and completeness of the underlying data before acting on it. Acting on inaccurate or incomplete data can expose companies to claims that termination was wrongful, exercised in bad faith, or unsupported by actual contractual defaults. Establish clear protocols for validating data before making termination decisions.

Assess Warranty, Quality, and Product Liability Considerations

Where digital twins inform process controls and quality thresholds, manufacturers should verify the reliability of the data inputs and models. Decisions based on flawed data can result in manufacturing defects, recalls, or regulatory findings. Clearly document the allocation of responsibility between themselves, contract manufacturers, and upstream suppliers to avoid disputes over liability.

Prepare for Regulatory and Audit Scrutiny

As regulators increasingly focus on AI governance and traceability, manufacturers should ensure that how digital twin outputs are used operationally aligns with how compliance obligations are documented. In regulated manufacturing sectors, inconsistencies between operational use and documented compliance can create significant exposure. Regular internal audits and documentation reviews can help identify and address these gaps before they become the subject of regulatory inquiry.

Positioning Your Organization to Mitigate Digital Twin and Predictive Analytics Risks

Digital twins and predictive analytics are rapidly becoming foundational to manufacturing supply chain management. Manufacturers that take a proactive approach to data governance and contractual alignment will be better positioned to capture the benefits of these technologies while managing the associated legal risks. When disputes do arise, they will be resolved under contracts, not code—making thoughtful contracting essential.

Manufacturers that recognize these risks now—and proactively address them through thoughtful contracting and governance—will be far better positioned to realize AI’s benefits without inheriting its litigation fallout. Â鶹´«Ă˝ & Lardner’s Manufacturing, Supply Chain, and Artificial Intelligence teams are available to help organizations navigate this evolving intersection of AI technology and traditional supply chain dynamics. We welcome the opportunity to discuss how these issues may affect your company’s operations.

Subscribe to the series to get updates about new articles.