UpGuard AI contract — TermScout contract intelligence report

UpGuard Generative AI Amendment Review & Analysis

UpGuard TermScout AI certified contract badge

AI Certification (Beta) Overview:

When TermScout certifies a contract, it means that agreement has met our rigorous standards and made the below commitments to their customers as it relates to their AI product or services.

Receiving Termscout's AI Certification (Beta) means that the product is on the cutting edge of AI development and shows they're committed to establishing trust and transparency with their customers and being a responsible corporate citizen when it comes to their AI development and functionalities.

Termscout acknowledges that presently AI is something of a moving target, but at this point, we believe responsible vendors make these minimum commitments and are continuously evaluating their practices.

AI Commitments Overview:

Review the AI-related commitments and view source language citations below.

UpGuard has established guardrails related to the use of their AI offering.
Status
UpGuard addresses how they use Customer data related to their AI offering.
Status
UpGuard does not address how they retain Customer data as it relates to their AI offering.
Status
Customer retains ownership of input data submitted to UpGuard as it relates to their AI offering.
Status
UpGuard does not commit to notifying Customer of a security breach impacting the Customer's data.
Status
UpGuard commits to complying with applicable AI laws
Status
In conversations between TermScout and UpGuard, UpGuard has not indicated their ongoing commitment to adopting contractual best practices related to AI as they're developed.
Status

AI Certification (Beta) Guiding Principles:

Termscout's AI Certification (beta) covers five major categories outlined below. Termscout believes while there are other important principles related to AI, these five categories cover the major obligations related to AI found in vendor contracts today and expect over time this will expand to fully realized compliance index.

1. Use Restrictions: Limiting harmful or unintended uses of AI Systems.
2. Transparency and Explainability: Ensuring users understand how AI works and its limitations.
3. Data Usage and Ownership: Clarifying rights around training data and generated content.
4. Security and Incident Response: Protecting data and systems from breaches or misuse.
5. Regulatory Compliance: Ensuring alignment with evolving laws.

AI Certification (Beta) Program:

If you are interested in being part of the AI Certification (Beta) Program and joining a standing group of select customers who has ongoing conversations with us about these standards and how they should evolve over time, please contact us at sales@termscout.com

Frequently Asked Questions

Find quick answers to the most common questions about our platform, process, and agreements.

Legal teams often escalate agreements when training provisions are broad enough to permit reuse of customer inputs, outputs, or operational data beyond service delivery. Escalation is common when the agreement lacks clear restrictions around retention, anonymization, subprocessors, or downstream commercial use. Counsel also scrutinizes language combining service-improvement rights with unrestricted model development authority.

Traditional analytics focus on aggregated optimization, while AI training rights can create long-term governance and competitive concerns beyond ordinary telemetry. Buyers evaluate whether customer data could influence shared models or generate outputs outside the intended relationship. Review therefore centers on scope limitation, reuse boundaries, and operational safeguards.

Agreements are viewed more favorably when they distinguish operational service functionality from broader model development. Enterprises respond well to narrowly tailored rights, explicit exclusions for confidential data, and transparent processing descriptions. Vendors reduce escalation pressure by explaining retention, permitting opt-outs, and avoiding ambiguous improvement language.

Counsel increasingly evaluates AI training clauses through a governance and downstream-exposure lens rather than as isolated IP provisions. Review includes regulatory obligations, auditability, data lineage, and reputational risk if sensitive data enters shared training environments. Concern rises when agreements do not explain dataset segregation, deletion handling, or whether customer content could persist in future model behavior.