Two years ago, many technology agreements addressed artificial intelligence (AI), if at all, through a generic disclaimer or a brief acknowledgment that AI features might be included in the offering. Today, that approach is inadequate. The integration of AI into commercial products, outsourcing arrangements, and enterprise software agreements has forced a rethinking of longstanding contract frameworks.
The SaaS Framework Is Under Pressure
Technology agreements for AI products have largely been structured as software-as-a-service (SaaS) arrangements: the customer pays for platform access, the vendor maintains the model, liability is capped, and the customer bears responsibility for how it uses the tool. That framework made reasonable sense when AI was a passive capability with a human making all consequential decisions.
However, the emergence of agentic AI and deeply embedded AI functionality is creating tension as the realities of “sufficient human intervention” become more apparent. We are seeing sophisticated buyers push for service-oriented terms: defined service descriptions, performance-based warranties, governance and audit rights, and outcome-tied liability structures. Vendors are resisting some of these provisions, but the negotiating dynamic is slowly shifting, at least in deals where AI performs critical enterprise tasks.
Key Negotiating Points
Scope and Service Definitions
Where AI performs tasks on a customer's behalf, the service description needs to articulate what the AI is actually doing—the workflows it executes, the decisions it makes, the systems it touches, and the guardrails that apply. Vague scope definitions create gaps that both parties will regret when something goes wrong.
Performance Standards and Warranties
Standard SaaS agreements for AI products frequently include broad disclaimers discarding any warranty as to accuracy or fitness for purpose. Where AI is embedded in core business processes and outcomes matter, buyers are increasingly unwilling to accept a warranty-free posture. We are seeing more deals include commitments around performance benchmarks, accuracy thresholds (which, although time intensive to create, are increasingly important), and obligations to remediate when outputs fall short.
Intellectual Property Ownership and Output Rights
Intellectual property (IP) allocation in AI agreements has become one of the most actively negotiated areas. Key issues to consider include who owns AI-generated outputs, what rights the customer retains in training data it provides, and whether the vendor can use customer data to improve its model. We recommend addressing output ownership, training data use limitations, and data return and deletion rights explicitly—tailored to the actual use case rather than carried over from a generic SaaS template.
Regulatory Compliance Allocation
The regulatory landscape for AI varies widely—from the EU AI Act, to US state-level requirements, such as those in Colorado, to federal guidance in financial services and healthcare. Agreements should address which party bears responsibility for compliance, how obligations are monitored, and how liability is allocated for regulatory penalties arising from the AI's use. Neither party benefits from ambiguity.
Transition and Termination
End-of-term planning should address continued access during wind-down, extraction of fine-tuned model components or proprietary configurations, return or deletion of customer data, and transition assistance. These provisions are often negotiated with less attention than they deserve—until a customer seeks to change vendors.
Avoid the Generic AI Disclaimer Trap
The better approach for both buy-side and sell-side is to integrate AI risk into the substance of the contract, including in the IP provisions, the liability framework, the data governance terms, the compliance allocation, and the service-level commitments. A well-drafted AI agreement reflects the specific use case and the actual risk profile of the deal. While it may require more work upfront, it produces an agreement that functions when it matters.
How We Can Help
Morgan Lewis's technology transactions, outsourcing, and commercial contracts lawyers regularly advise clients on both the buy-side and sell-side in negotiating complex technology agreements with significant AI components, including enterprise software agreements, outsourcing arrangements, and AI-specific transactions. If you have questions about the topics discussed above or would like to learn more, please reach out to any member of our team.