Responsible Use of Artificial Intelligence Policy for Bitsight Vendors

1. Introduction

This Responsible Use of Artification Intelligence Policy for Bitsight Vendors (the “Policy”) applies to all vendors, partners, contractors, and third-party agents (collectively referred to as “Vendor”) engaged by BitSight Technologies, Inc. and its subsidiaries (collectively “Bitsight”). It is designed to ensure that Vendor adheres to the same standards that Bitsight expects of its own employees, consultants, and contractors with respect to the use of machine learning and AI tools (e.g. ChatGPT, Google AI Studio/Bard, GitHub CoPilot, Claude, and others) (collectively “AI Tools”).

2. Governing Principles for AI Tool Usage

Vendors are expected to only use AI Tools for lawful and ethical purposes. Accordingly, Vendor is prohibited from using AI Tool to:

  • Facilitate or promote illegal activities or violations of law;
  • Access or create harmful, offensive, discriminatory, or inappropriate content;
  • Attempt to override or circumvent safety filters (except in cases where individual employees have been explicitly authorized by Bitsight’s AI Council to do so for the purpose of Bitsight’s security research efforts) or intentionally drive the AI Tool to act in a manner that contravenes Bitsight policies that Bitsight has made Vendor aware of or that Vendor should otherwise be aware of in its course of dealings with Bitsight;
  • Misrepresent the origin of AI-generated content by claiming content was created by a particular human;
  • Make automated decisions relating related to natural persons without human involvement (e.g. employment decision-making, profiling);
  • Infringe the rights of others, including privacy and intellectual property rights; or
  • Knowingly distribute false or misleading information.

3. Usage Guidelines:

Vendor must ensure that its use of AI Tools complies with all applicable laws and regulations, including those concerning intellectual property, privacy, and data protection. Vendor at all times must safeguard sensitive, confidential, and proprietary information belonging to Bitsight, its employees, customers, or other third parties. Vendor must not input such information into AI Tools without express written authorization from Bitsight. Further, Vendor must ensure that any AI generated content it is using in the Services or providing to Bitsight in any way as part of the services is factually accurate, and aligns with Bitsight’s values, ethics, and quality standards as set forth in Bitsight’s Vendor Code of Conduct.

To this end, when using AI Tools in the workplace, Vendor users must:

  • Thoroughly review all AI Tool outputs before using them or forwarding them to others inside or outside Bitsight to: (i) confirm that they do not contain biased, offensive, or discriminatory content; (ii) confirm they do not improperly use or disclose personal or confidential information; and (iii) verify accuracy or reported facts with other trusted sources.
  • Not enter any Bitsight employee, customer, or third-party sensitive, confidential, trade secret, or other personal or proprietary information into a prompt for an AI Tool unless expressly authorized in advance by Bitsight.
  • Not use any AI Tools for creating or assisting in creating any Bitsight products or services where Bitsight will seek copyright or patent protection without first obtaining explicit advance permission from Bitsight by emailing [email protected].
  • Not use free or public facing versions of AI Tools for the services it is providing to Bitsight.

Vendor is responsible for all work product its users create using AI Tools and will be held accountable for such output. Vendor is also responsible for all input used with AI Tools including but not limited to the examples provided in this policy. Bitsight retains the right to audit Vendor’s use of AI Tools in accordance with the agreement for services between Bitsight and Vendor.

Vendor must report any non-compliance with this Policy to Bitsight’s Information Security team at [email protected] and adhere to the corrective actions as directed. Further, Vendor agrees that it will ensure that all of its employees, contractors, and other parties using AI Tools in connection with providing services to Bitsight are aware of and comply with this Policy or other restrictions and requirements that are substantially similar to the restrictions and requirements contained in this Policy.

4. Enforcement and Compliance

Failure to comply with this Policy is considered a breach of the agreement between Bitsight and the Vendor and may result in termination of the Vendor’s engagement with Bitsight, as well as other remedies available under contract or law. If Vendor becomes aware of an actual or potential violation of this Policy, or has reason to believe that a violation of this Policy has occurred, it must promptly disclose this fact to Bitsight to [email protected] together with all relevant documents and information. Violations of this Policy include but are not limited to the following:

  • Use of an unlicensed AI Tool, if usage requires a license.
  • An AI Tool being used outside of the approved manner or scope, including in contradiction of this Policy.
  • Use of an AI Tool that poses an identified, unaddressed security risk or contains any material defects or malicious code.
  • Publication of inaccurate or misleading information on behalf of Bitsight as a result of use of an AI Tool.

Bitsight prohibits any form of discipline, reprisal, intimidation, or retaliation for reporting a violation of this Policy.

5. Summary and Questions

Bitsight reserves the right to change, modify, or delete provisions of this Policy without notice. Bitsight’s Information Security Department is responsible for the administration of this Policy. For all inquiries about this Policy, please email Bitsight’s Information Security team at [email protected].