It’s Thanksgiving, and the turkey’s in the oven. The perfect feast requires attentiveness to cooking the Holiday dishes, pouring the right wine, and setting a memorable table. If you’re off by even a little, the bird, if not the entire dinner, may be at risk. Certainly, preparing a Thanksgiving dinner is akin to running a business – success, or failure, depends on coordinating many moving parts. Just like a Thanksgiving meal comes together with contributions from every chef and dish, your business thrives when its people and parts work in harmony.
In Part 2 of our four-part blog series on Artificial Intelligence (“AI”) and ERP technology, we further explore the unique financial and legal perils triggered by AI which put your business at risk. Enterprise Resource Planning (ERP) systems are ubiquitous in business operations, and the integration of AI into ERP systems has made these platforms more powerful and more likely to create significant financial and legal risks – especially in commercial transactions such as mergers and acquisitions (M&A).
M&A deals have long faced legal complexities related to ERP software licenses, including assignment restrictions which stir issues of transferability, compliance, and cost. Now, thanks to the rise of AI, these challenges have been exacerbated, making pre-deal due diligence and post-deal integration even more thorny.
7 Key Financial & Legal Perils of Commercial Transactions Involving AI-Enabled ERP Systems
- Licensing and transferability. AI modules within ERP systems often come with licensing restrictions that may not align with M&A transactions. Some ERP licenses will not allow assignments or transfers, while free or limited-use versions may impose risky terms, such as indemnity obligations or vendor ownership of outputs. Post-transaction, the buyer may incur significant costs to renegotiate licenses, upgrade to compliant versions, or replace non-transferable AI components.
- Vendor obligations and indemnities. Some AI-enabled ERP contracts include clauses requiring the customer to indemnify the vendor for liabilities arising from the use of the AI system. If the target company’s indemnity obligations are transferred to the buyer, the buyer may be exposed to significant financial liabilities, particularly if the AI outputs cause harm or infringe on third-party rights.
- Operational dependencies. ERP systems with integrated AI are often deeply embedded in a company’s operations. The target company's reliance on proprietary AI systems can make integration with the buyer’s systems difficult. Post-transaction, the buyer may face operational inefficiencies, increased costs, or disruptions if the AI systems cannot be integrated or require significant modification.
- Security and confidentiality risks. AI-driven ERPs process sensitive business information. Some AI modules, particularly those connected to cloud services, may not guarantee confidentiality of inputs and outputs. Breaches of confidentiality can lead to loss of trade secrets, competitive advantage, or regulatory penalties.
- Algorithmic bias and liability. AI in ERP systems can inadvertently produce biased or discriminatory outcomes (e.g., in recruitment or customer segmentation), especially if the training data was flawed. The buyer could face lawsuits, regulatory investigations, or reputational harm from biased decisions made by the AI.
- Data ownership and usage rights. AI-enabled ERPs depend on large datasets for functionality, including customer data, financial records, and supply chain metrics. In commercial transactions, determining whether the target company has legal ownership or appropriate rights to this data is critical. If data was improperly obtained or used for training AI models, the buyer could face regulatory penalties or be forced to delete critical data, disrupting operations. Violations of data privacy laws could result in fines and loss of business value.
- Regulatory compliance. AI-enabled ERPs often handle sensitive operations, such as employment decisions or consumer finance. These activities are increasingly regulated under laws such as United States AI legislation or the EU’s AI Act. Non-compliance can result in fines, mandatory operational changes, or reputational damage. The buyer may inherit these liabilities during an acquisition.
Mitigation Strategies
The presence of AI on the menu means taking your due diligence and best practices to another level of care and scrutiny. While anticipating the entirety of unforeseen risk and peril may be next to impossible, there are steps that can be taken at the outset to mitigate risk.
AI-Specific Due Diligence. Expand due diligence to include:
- Detailed review of AI-related licenses and terms.
- Assessment of data ownership and compliance with data privacy laws.
- Evaluation of the AI's regulatory compliance, especially in sensitive areas like employment or consumer finance.
Engage AI and Legal Experts. Include professionals familiar with AI law, ERP technologies, and licensing practices to identify and address risks effectively.
Negotiate Protections in Transaction Documents. Incorporate AI-specific representations, warranties, and indemnities into purchase agreements to protect against undisclosed liabilities.
Develop an Integration Plan. Create a strategy for integrating AI-enabled ERP systems post-transaction, including addressing licensing gaps and operational dependencies.
Monitor Regulatory Changes. Stay informed about evolving AI regulations to ensure compliance and mitigate future liabilities.
In a manner of speaking, you need a lot of chefs in the kitchen when addressing and managing AI risks in your ERP software licenses. Finding the right professionals and paying attention to what’s on the stove is the best way to assure that you don’t become toast when managing your AI affairs.
Published on November 27, 2024
