Proposed US Social Media Disclosure and Transparency of Advertisements Act of 2021

Government policy documents

Website/link: https://trahan.house.gov/uploadedfiles/social_media_data_act_section_by_section.pdf

Website/link: Visit website

Published date: May 25, 2021

Author: Lori Trahan

Country: United States of America (USA)

Legislation status: Draft legislation

Subject tag: Advertising | Data Access | Privacy and data protection

This section establishes a Working Group for Social Media Research Access that will
• Offer input into a Best Practices and Code of Conduct Report, which will be published by the FTC and includes:
o A description of the best practices regarding what type of information from an interactive computer service should be made available, and under what circumstances, to academic researchers.
o A recommended code of conduct for academic researchers working with such information. • Provide Policy Recommendations to Congress on:
o The types of information and under what circumstances interactive computer services should be required to grant access to academic researchers.
o Circumstances in which additional legal protections for interactive computer services may be needed related to the sharing of data with researchers.
o Recommendation for penalties for academic researchers who misuse or seek to inappropriately reidentify information provided to them by interactive computer services for research purposes.
This section also outlines a list of considerations for the working group to make while drafting
the reports.
• The type of information considered, including (but not limited to) the following:
o Information related to content moderation decisions.
o Information related to engagement.
o Information related to exposure.
o Classification of public information sources.
o Archives of formerly public accounts that were removed.
o Archives of fake or bot accounts that have been removed.
o Archives of coordinated influence operation accounts that have been removed.
o Internal research conducted by an interactive computer service.
o The most popular content on a platform.
• Storage of information and treatment for each type of information considered, including (but not limited to) the following:
o Limits on time and amount of information stored broken down by the type of information.
o Under what circumstances privacy preserving techniques such as differential privacy and statistical noise could be used.
o Required level of aggregation for demographic information.
o Standardized variable names across platforms for specific types of information.
o Under what circumstances erasure policies (related to who and how an individual can request to be removed from a dataset) may be needed, specifically for individuals who are less than 18 years old.
o Adherence to data security best practices.
• A consideration of access for each type of information considered, including (but not limited to) the following:
o Designation of secure facilities and computers to analyze information.
o Set criteria for researcher access.
o Any limit on the type of research that specific datasets can be used for.
o An analysis of how any of the recommendations might interact with international law and jurisdiction.
o Policies for assuring that open science principles of reproducibility of results and replication of analyses can be respected.
o Recommendations for the size and type of interactive computer service, which may vary. Section 2(c )
[This entry was sourced with minor edits from the Carnegie Endowment’s Partnership for Countering Influence Operations and its baseline datasets initiative. You can find more information here: https://ceip.knack.com/pcio-baseline-datasets]