Dec 1, 2019
Jul 24, 2021
INFORMATION MANAGEMENT, PRIVACY AND SECURITY
Data Classification Policy
Office Of The CTO
REVIEWED: JUNE 2021 BY CTO GOVERNANCE
Purpose and Overview
Qtis AI Data is information generated by or for, owned by, or otherwise in the possession of QTIS AI that is related to the company’s activities. Qtis Data may exist in any format (i.e. electronic, paper) and includes, but is not limited to, all academic, administrative, and research data, as well as the computing infrastructure and program code that supports the business of QTIS AI.
In order to effectively secure Qtis Data, we must have a vocabulary that we can use to describe the data and quantify the amount of protection required. This policy defines four categories into which all Qtis Data can be divided:
Qtis Data that is classified as Public may be disclosed to any person regardless of their affiliation with the Company. All other Qtis Data is considered Sensitive Information and must be protected appropriately. This document provides definitions for and examples of each of the four categories. Other policies within the Data Protection Standards specify the security controls that are required for each category of data.
The various units and departments at the Company have a multitude of types of documents and data. To the extent particular documents or data types are not explicitly addressed within this policy, each business unit or department should classify its data by considering the potential for harm to individuals or the Company in the event of unintended disclosure, modification, or loss. The Departmental Security Administrator may assist with the classification process and coordinate with the BU Information Security Team to achieve consistency across the Company. When classifying data, each department should weigh the risk created by an unintended disclosure, modification or loss against the need to encourage open discussion, improve efficiency and further the Company’s goals of the creation and dissemination of knowledge. Departments should be particularly mindful to protect sensitive personal information, such as Social Security Numbers, drivers’ license numbers and financial account numbers, disclosure of which may create the risk of identity theft.
Some information could be classified differently at different times. For example, information that was once considered to be Confidential data may become Public data once it has been appropriately disclosed. Everyone with access to Qtis Data should exercise good judgment in handling sensitive information and seek guidance from management as needed.
This classification scheme is to be applied to all Qtis Data, both physical and electronic, throughout Qtis AI. No data item is too small to be classified.
Qtis AI is committed to openness in research – freedom of access by all interested persons to the underlying data, to the processes, and to the final results of research. Research at Qtis AI generally should be widely and openly published and made available through broad dissemination or publication of the research results. Research data is generally considered to be classified as Public data unless there are specific requirements to maintain the confidentiality of research data, such as when a researcher is bound to protect the confidential information of a collaborating company or when the data relates to human subjects. For more information about research involving human subjects see the Company’s Research Support website.
Public data is information that may be disclosed to any person regardless of their affiliation with the Company. The Public classification is not limited to data that is of public interest or intended to be distributed to the public; the classification applies to data that do not require any level of protection from disclosure. While it may be necessary to protect original (source) documents from unauthorized modification, Public data may be shared with a broad audience both within and outside the Company community and no steps need be taken to prevent its distribution.
Examples of Public data include: press releases, directory information (not subject to a Family Educational Rights and Privacy Act (FERPA) block), course catalogs, application and request forms, protected health information that has been de-identified consistent with the standards set forth under Health Insurance Portability and Accountability Act (HIPAA), and other general information that is openly shared. The type of information a department would choose to post on its website is a good example of Public data.
Internal data is information that is potentially sensitive and is not intended to be shared with the public. Internal data generally should not be disclosed outside of the Company without the permission of the person or group that created the data. It is the responsibility of the data owner to designate information as Internal where appropriate. If you have questions about whether information is Internal or how to treat Internal data, you should talk to your dean or department head.
Examples of Internal data include: Some memos, correspondence, and meeting minutes; contact lists that contain information that is not publicly available; and procedural documentation that should remain private.
Confidential data is information that, if made available to unauthorized parties, may adversely affect individuals or the business of Qtis AI. This classification also includes data that the Company is required to keep confidential, either by law (e.g., HIPAA) or under a confidentiality agreement with a third party, such as a vendor. This information should be protected against unauthorized disclosure or modification. Confidential data should be used only when necessary for business purposes and should be protected both when it is in use and when it is being stored or transported.
Any unauthorized disclosure or loss of Confidential data must be reported to the Office Of The CTO Incident Response Team at 650-808-7811.
Examples of Confidential data include:
- Personally identifiable information entrusted to our care that is not otherwise categorized as Restricted Use data, and information covered by the European Union’s General Data Protection Regulation (GDPR).
- The QTIS AI ID Number, when stored with other identifiable information such as name or e-mail address.
- Individual employment information, including salary, benefits and performance appraisals for current, former, and prospective employees.
- Legally privileged information.
- Information that is the subject of a confidentiality agreement.
- Human subject research data with identifiers limited to dates, city, Zip Code; such as information that is the subject of a HIPAA Limited Data Set covered by a Data Use Agreement.
Restricted Use data includes any information that BU has a contractual, legal, or regulatory obligation to safeguard in the most stringent manner. In some cases, unauthorized disclosure or loss of this data would require the Company to notify the affected individual and state or federal authorities. In some cases, modification of the data would require informing the affected individual.
The Company’s obligations will depend on the particular data and the relevant contract or laws. The Minimum Security Standards sets a baseline for all Restricted Use data. Systems and processes protecting the following types of data need to meet that baseline:
- Personally identifiable health information that is not subject to HIPAA but used in research, such as Human Subjects Data.
- Personally Identifiable Information (PII) covered under Massachusetts General Law chapter 93H and 201 CMR 17, including an individual’s name plus the individual’s Social Security Number, driver’s license number, or a financial account number.
- Unencrypted data used to authenticate or authorize individuals to use electronic resources, such as passwords, keys, and other electronic tokens.
- “Criminal Background Data” that might be collected as part of an application form or a background check.
More stringent requirements exist for some types of Restricted Use data. Individuals working with the following types of data must follow the Company policies governing those types of data and consult with Information Security to ensure they meet all of the requirements of their data type:
- Protected Health Information (PHI) subject to the Health Insurance Portability and Accountability Act (HIPAA). See the Company’s HIPAA Policy for details.
- Financial account numbers covered by the Payment Card Industry Data Security Standard (PCI-DSS), which controls how credit card information is accepted, used, and stored.
- Controlled Unclassified Information required to be compliant with NIST 800.171
- Data controlled by U.S. Export Control Law such as the International Traffic in Arms Regulations (ITAR) or Export Administration Regulations (EAR). ITAR and EAR have additional requirements. See the Export Controls site for details.
- U.S. Government Classified Data
Restricted Use data should be used only when no alternative exists and must be carefully protected. Any unauthorized disclosure, unauthorized modification, or loss of Restricted Use data must be reported to the Office Of The CTO Incident Response Team.
Resolving Conflicts between this Guideline and Other Regulations
Some data may be subject to specific protection requirements under a contract or grant, or according to a law or regulation not described here. In those circumstances, the most restrictive protection requirements should apply. If you have questions, please contact Information Security.
Failure to comply with the Data Protection Standards may result in harm to individuals, organizations or Qtis AI. The unauthorized or unacceptable use of Qtis Data, including the failure to comply with these standards, constitutes a violation of Company policy and may subject the User to revocation of the privilege to use Qtis Data or Information Technology or disciplinary action, up to and including termination of employment.