Vivek Rao, Kyle Glover, and Matt Stein Author Global Legal Post Article: Negotiating Bespoke AI Contract Terms: Moving Beyond Boilerplate
As businesses increasingly adopt AI technologies, AI contracting is moving beyond the “AI Terms 1.0” approach of competing boilerplate positions toward more tailored, risk-informed negotiations.
In their Artificial Intelligence, Law Over Borders Comparative Guide 2026 chapter, Pierce Atwood attorneys Vivek Rao, Kyle Glover, and Matthew Stein examine this shift and explain how vendors and customers can negotiate AI contracts based on the technology’s specific use case and associated risks.
The authors describe the earlier contracting landscape as largely divided between vendors seeking broad data-use rights, extensive disclaimers, and limited indemnification, and sophisticated customers demanding extensive warranties, documentation, audit rights, and broader indemnification. Today, negotiations between parties with comparable sophistication and leverage increasingly reflect a more nuanced approach, with each side identifying the risks it is best positioned to manage and negotiating targeted protections accordingly.
The article recommends conducting a focused AI risk assessment before negotiations begin. Among other considerations, parties should identify who could be harmed, the nature of potential harms, which party is best positioned to prevent or mitigate those harms, and what technical, administrative, and contractual safeguards are appropriate for the particular use case. The resulting assessment can help parties distinguish material risks that warrant firm contractual protections from lower-risk issues that may be better addressed through operational controls or reasonable compromises.
The authors explore several areas where this risk-informed approach can produce more practical contract terms, including:
- Intellectual property ownership
- Secondary data use
- Model documentation
- Output quality and fairness
- Regulatory compliance
- Infringement risk
- Disclaimers and liability
- Model changes and monitoring
The authors conclude that effective AI contracting does not require either side to achieve every theoretically ideal protection. Instead, parties should focus on the risks that matter most for the particular AI application and negotiate meaningful safeguards around those risks. This tailored approach can help businesses capture the benefits of AI while establishing practical mechanisms to manage its evolving legal and operational risks.
“Negotiating Bespoke AI Contract Terms” appears in Global Legal Post’s Artificial Intelligence, Law Over Borders Comparative Guide 2026