Considerations for AI SaaS Agreements

Many firms are integrating AI by working with outside vendors. Oftentimes, these deals are structured as Software as a Service (SaaS) agreements. This usually looks like a firm paying a monthly fee to a vendor for access to that vendor’s AI systems. While these agreements may seem straightforward, individual needs and scale can quickly complicate agreements. A recent Venable memo gives tips and key considerations to keep in mind while negotiating AI SaaS agreements. Here are a couple of highlights:

“Clarify the Scope of the AI Services

• Ensure the agreement defines what the AI system actually does, including use cases, functionality, and any limitations
• Look for ambiguous descriptions like “predictive” or “autonomous,” which can inflate expectations and complicate performance disputes
• Verify whether the provider uses third-party models, APIs, or datasets, as this may affect risk allocation and licensing rights

Address Confidentiality and AI Output Handling

• Ensure that AI outputs—especially those containing or derived from confidential data—are expressly protected under the confidentiality clause
• Require the provider to implement technical and organizational safeguards to prevent unintended disclosure or model leakage
• Consider contract terms limiting the provider’s ability to use or reuse outputs for other customers, where feasible.”

Negotiating your SaaS agreements ensures that your company’s needs are being met with proper functionality, security, and data management.