# Data Dictionary This document outlines metadata for each column in the OMB-consolidated individual AI use case Inventory. > **Encoding Notice:** CSV files use `UTF-8-SIG` encoding for compatibility with special characters (e.g., em dashes). For programmatic analysis, specify `utf-8-sig` to ensure correct character rendering. > **List Formatting:** The 'Select Multiple' columns (`demographic_features` and `hi_public_consultation`) are stored as stringified Python lists (e.g., `['Value 1', 'Value 2']`). For analysis, these can be parsed using standard list evaluation or JSON loading tools. **Note**: Recodings were done for multiple choice columns when reported answers had slight spelling/formatting deviations from the valid selections and were clearly intended to match one of the valid options (e.g. mapping 'false' to 'no' or 'None' to 'None of the above'). For responses that were not similar to any of the valid options, no changes were made except for stripping excess leading/trailing whitespace (except for a small handful of occurrences in the development stage column). Lists were reformatted to use a semicolon separator and to be sorted in alphabetical order. See the `data_standardization_report.md` file for more details. --- ## Individual Use Case Inventory - Column Metadata ### 1) `agency` *Note:* This column was generated by OMB - **Data Type:** `free_text` - **Instructions:** The abbreviated name of the reporting agency (e.g., USDA). --- ### 2) `agency_name` *Note:* This column was generated by OMB - **Data Type:** `free_text` - **Instructions:** The full name of the reporting agency (e.g., The Department of Agriculture). --- ### 3) `id` - **Original Excel Header:** 'Use Case ID [Agency Abbrev.] – [#]' - **Data Type:** `structured_text` - **Instructions:** 'Unique identifier for AI use case. Use cases should maintain the same ID if included in previous year’s inventory. However, in their public inventories, agencies are permitted to remove the “Use Case ID” column. Required: Yes' --- ### 4) `use_case_name` - **Original Excel Header:** 'Use Case Name' - **Data Type:** `free_text` - **Instructions:** 'Title of the AI use case. Required: Yes' --- ### 5) `agency_bureau` - **Original Excel Header:** 'Bureau/Component' - **Data Type:** `free_text` - **Instructions:** 'Specific organization(s) within the agency responsible for the AI use case. Required: Yes' --- ### 6) `contact_email` - **Original Excel Header:** 'Email Address' - **Data Type:** `free_text` - **Instructions:** 'Email address of the individual or team who works with or is responsible for the AI use case. However, in their public inventories, agencies may replace the “Email Address” column with an email address designated for public inquiries. Required: Yes' --- ### 7) `is_withheld` - **Original Excel Header:** 'Should this AI use case be withheld from public reporting?' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether the AI use case should be withheld from public reporting and the reason for why. Required: Yes' **Valid Selections:** | Valid Value | | :--- | | a) No | | b) Yes – agency has determined that there’s a risk to disclosure, such as a harm to an interest protected by a FOIA exemption | | c) Yes – disclosure is prohibited by law | | d) Other | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) No | **No** | | b) Yes – agency has determined that there’s a risk to disclosure, such as a harm to an interest protected by a FOIA exemption | **Yes - Disclosure Risk** | | c) Yes – disclosure is prohibited by law | **Yes - Prohibited by Law** | | d) Other | **Other** | --- ### 8) `development_stage` - **Original Excel Header:** 'Stage of Development' - **Data Type:** `multiple_choice` - **Instructions:** 'The AI use case’s current stage of development. Required: Yes' **Valid Selections:** | Valid Value | | :--- | | a) Pre-deployment – The use case is in a development or acquisition status. | | b) Pilot – The use case has been deployed in a limited test or pilot capacity. | | c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency. | | d) Retired – The use case was reported in the agency’s prior year’s inventory, but its development and/or use has since been discontinued. | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Pre-deployment – The use case is in a development or acquisition status. | **Pre-deployment** | | b) Pilot – The use case has been deployed in a limited test or pilot capacity. | **Pilot** | | c) Deployed – The use case is being actively authorized or utilized to support the functions or mission of an agency. | **Deployed** | | d) Retired – The use case was reported in the agency’s prior year’s inventory, but its development and/or use has since been discontinued. | **Retired** | --- ### 9) `is_high_impact` - **Original Excel Header:** 'Is the AI use case high-impact?' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether the AI use case is high-impact, as defined in Section 5 of the appendix to OMB Memorandum M-25-21. Required: Yes' **Valid Selections:** | Valid Value | | :--- | | a) High-impact | | b) Presumed high-impact, but determined not high impact | | c) Not high-impact | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) High-impact | **High-impact** | | b) Presumed high-impact, but determined not high impact | **Presumed High-Impact, but Not High-impact** | | c) Not high-impact | **Not High-impact** | --- ### 10) `HI_justification` - **Original Excel Header:** 'Justification' - **Data Type:** `free_text` - **Instructions:** 'Provide a justification for why the use case is determined to be not high-impact. Required: Yes, but question should only appear if “Presumed high-impact, but determined not high-impact” is selected for the previous question.' --- ### 11) `topic_area` - **Original Excel Header:** 'Use Case Topic Area' - **Data Type:** `multiple_choice` - **Instructions:** 'The topic area that most closely aligns with the AI use case. Required: Yes, for pre-deployment, pilot and deployed use cases.' **Valid Selections:** | Valid Value | | :--- | | Administrative Functions | | Cybersecurity | | Emergency Management | | Energy and the Environment | | Government Benefits Processing | | Health and Medical | | Human Resources | | Information Technology | | International Affairs | | Law Enforcement | | Other | | Procurement and Financial Management | | Science | | Service Delivery | | Transportation | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | Administrative Functions | **Admin Functions** | | Cybersecurity | **Cybersecurity** | | Emergency Management | **Emergency Mgmt** | | Energy and the Environment | **Energy and Environment** | | Government Benefits Processing | **Benefits Processing** | | Health and Medical | **Health and Medical** | | Human Resources | **HR** | | Information Technology | **IT** | | International Affairs | **International Affairs** | | Law Enforcement | **Law Enforcement** | | Other | **Other** | | Procurement and Financial Management | **Procurement and Finance Mgmt** | | Science | **Science** | | Service Delivery | **Service Delivery** | | Transportation | **Transportation** | --- ### 12) `classification` - **Original Excel Header:** 'AI Classification' - **Data Type:** `multiple_choice` - **Instructions:** 'The classification of AI technology that most closely aligns with the AI use case. Note that an AI use case may rely on multiple classifications; however, agencies should select the choice that most closely aligns. Required: Yes, for pre-deployment, pilot and deployed use cases.' **Valid Selections:** | Valid Value | | :--- | | Agentic AI: AI systems that perform tasks or make decisions autonomously with minimal human intervention. | | Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relationships. | | Computer Vision: AI that processes and interprets visual data (e.g., images and videos). | | Generative AI: AI that generates new or synthetic content (e.g., images, videos, audio, text, code). | | Natural Language Processing: AI that processes, interprets, and shares information in human language. | | Reinforcement Learning: AI trained through trial and error using rewards and penalties to optimize decision-making policies. | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | Agentic AI: AI systems that perform tasks or make decisions autonomously with minimal human intervention. | **Agentic AI** | | Classical/Predictive Machine Learning: Models trained on data to make predictions or classifications based on identified patterns or relationships. | **Classical ML** | | Computer Vision: AI that processes and interprets visual data (e.g., images and videos). | **Computer Vision** | | Generative AI: AI that generates new or synthetic content (e.g., images, videos, audio, text, code). | **Generative AI** | | Natural Language Processing: AI that processes, interprets, and shares information in human language. | **NLP** | | Reinforcement Learning: AI trained through trial and error using rewards and penalties to optimize decision-making policies. | **Reinforcement Learning** | --- ### 13) `problem_solved` - **Original Excel Header:** 'What problem is the AI intended to solve?' - **Data Type:** `free_text` - **Instructions:** 'What problem is the AI intended to solve? Required: Yes, for pre-deployment, pilot and deployed use cases.' --- ### 14) `benefits` - **Original Excel Header:** 'What are the expected benefits and positive outcomes from the AI for an agency’s mission and/or the general public?' - **Data Type:** `free_text` - **Instructions:** 'Description of the expected benefits and positive outcomes from the AI for an agency’s mission and/or the general public. Expected benefits may include, for example: cost savings, reducing customer wait times, or improving access to government benefits. Required: Yes, for pre-deployment, pilot and deployed use cases.' --- ### 15) `system_outputs` - **Original Excel Header:** 'Describe the AI system’s outputs.' - **Data Type:** `free_text` - **Instructions:** 'Description of what the AI system outputs, whether it’s a prediction, recommendation, decision, etc. Required: Yes, for pre-deployment, pilot and deployed use cases.' --- ### 16) `operational_date` - **Original Excel Header:** 'Date when AI use case became operational or the pilot’s start date' - **Data Type:** `date` - **Instructions:** 'Date when the AI use case was fully operational or the pilot’s start date. Required: Yes, for pilot and deployed use cases.' --- ### 17) `contracting_usage` - **Original Excel Header:** 'Was the system involved in this use case purchased from a vendor or developed under contract(s) or in-house?' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether the system involved in the use case was developed, or is expected to be developed, exclusively with contracting resources, in-house, or a combination of both. Required: Yes, for pilot and deployed use cases.' **Valid Selections:** | Valid Value | | :--- | | a) Purchased from a vendor | | b) Developed in-house | | c) Developed with both contracting and in-house resources | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Purchased from a vendor | **Vendor Purchased** | | b) Developed in-house | **In-house Development** | | c) Developed with both contracting and in-house resources | **Contracting and In House** | --- ### 18) `vendor_name` - **Original Excel Header:** 'Vendor(s) Name' - **Data Type:** `free_text` - **Instructions:** 'Identification of the vendor(s) involved in the use case’s development and/or maintenance. Required: Yes, but question should only appear if “Purchased from a vendor” or “Developed with both contracting and in-house resources” is selected for the previous question and if development stage is pilot or deployed.' --- ### 19) `have_ato` - **Original Excel Header:** 'Does this AI use case have an associated Authorization to Operate (ATO)?' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether the AI use case itself has an associated Authorization to Operate (ATO), or is part of another system that has an ATO. Required: Yes, for pilot and deployed use cases.' **Valid Selections:** | Valid Value | | :--- | | No | | Yes | --- ### 20) `system_name_ato` - **Original Excel Header:** 'System(s) Name' - **Data Type:** `free_text` - **Instructions:** 'Name of the system(s) associated with this AI use case, according to the ATO. Required: Yes, but question should only appear if “Yes” is selected for the previous question and development stage is pilot or deployed.' --- ### 21) `data_description` - **Original Excel Header:** 'Describe any data used to train, fine-tune, and/or evaluate performance of the model(s) used in this use case.' - **Data Type:** `free_text` - **Instructions:** 'General description of the data used in the AI’s design, development, training, testing, and operation. This should include a description of any data provided by the agency, whether the datasets are research datasets, publicly available, external, etc. and to the extent possible, sufficient descriptive information from the vendor. Required: Yes, for pilot and deployed use cases.' --- ### 22) `link_to_data` - **Original Excel Header:** 'If the data is required to be publicly disclosed as an open government data asset, provide a link to the entry on the Federal Data Catalog.' - **Data Type:** `free_text` - **Instructions:** 'If the data is required to be publicly disclosed as an open government data asset, provide a link to the entry on the Federal Data Catalog. Required: No' --- ### 23) `has_pii` - **Original Excel Header:** 'Does this AI use case involve personally identifiable information (PII) that is maintained by the agency?' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether the AI use case involves personally identifiable information (PII), as defined in OMB Circular A-130. Required: Yes, for pilot and deployed use cases.' **Valid Selections:** | Valid Value | | :--- | | No | | Yes | --- ### 24) `pia_url` - **Original Excel Header:** 'If publicly available, provide the link to the AI use case’s associated Privacy Impact Assessment (PIA).' - **Data Type:** `free_text` - **Instructions:** 'Whether the AI use case itself has an associated PIA, or is part of another system that has a PIA. Required: No' --- ### 25) `demographic_features` - **Original Excel Header:** 'Which, if any, demographic variables does the AI use case explicitly use as model features?' - **Data Type:** `select_multiple` - **Instructions:** 'Identification of the demographic features explicitly utilized in the AI use case’s data/model(s). Required: Yes, for pilot and deployed use cases.' **Valid Selections:** | Valid Value | | :--- | | a) Race/Ethnicity | | b) Sex | | c) Age | | d) Religious Affiliation | | e) Socioeconomic Status | | f) Ability Status | | g) Residency Status | | h) Marital Status | | i) Income | | j) Employment Status | | k) None of the above | | l) Other | --- ### 26) `has_custom_code` - **Original Excel Header:** 'Does this project include custom-developed code?' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether this AI project includes “custom developed code,” as defined in OMB Memorandum M-25-21. Required: Yes, for pilot and deployed use cases.' **Valid Selections:** | Valid Value | | :--- | | No | | Yes | --- ### 27) `code_url` - **Original Excel Header:** 'If the code is open source, provide the link for the publicly available source code.' - **Data Type:** `free_text` - **Instructions:** 'Link to publicly available source code for the AI use case, if available. Required: No' --- ### 28) `hi_testing_conducted` - **Original Excel Header:** 'Has pre-deployment testing been conducted for this AI use case? Practice: Complete AI Impact Assessment' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether pre-deployment testing has been conducted for the AI use case, consistent with Section 4(b)(i) of the appendix to OMB Memorandum M-25-21. Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Yes | | b) In-progress | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Yes | **Yes** | | b) In-progress | **In-progress** | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | --- ### 29) `hi_assessment_completed` - **Original Excel Header:** 'Has an AI impact assessment been completed for this AI use case? Practice: Complete AI Impact Assessment' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether an AI impact assessment for the AI use case has been completed, consistent with Section 4(b)(ii) of the appendix to OMB Memorandum M-25- 21. Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Yes | | b) In-progress | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Yes | **Yes** | | b) In-progress | **In-progress** | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | --- ### 30) `hi_potential_impacts` - **Original Excel Header:** 'What are the potential impacts of using the AI for this particular use case and how were they identified? Subpractice: Complete AI Impact Assessment' - **Data Type:** `free_text` - **Instructions:** 'Identification of the reasonably foreseeable impacts from using the AI use case, to include impacts on the privacy, civil rights, and civil liberties of the public. Required: Yes, only for high-impact deployed use cases' --- ### 31) `hi_independent_review` - **Original Excel Header:** 'Has as independent review of the AI use case been conducted? Sub practice Complete AI Impact Assessment' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether an independent review of the AI use case has been conducted by an office, oversight board, or individual within the agency who has not been involved in the AI’s development.Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Yes – by another appropriate agency office or reviewer not directly involved in the AI’s development | | b) Yes – by an agency AI oversight board not directly involved in the AI’s development | | c) Yes – by the CAIO | | d) In-progress | | e) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Yes – by another appropriate agency office or reviewer not directly involved in the AI’s development | **Internal Independent Review** | | b) Yes – by an agency AI oversight board not directly involved in the AI’s development | **Oversight Board Review** | | c) Yes – by the CAIO | **CAIO Review** | | d) In-progress | **In-progress** | | e) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | --- ### 32) `hi_ongoing_monitoring` - **Original Excel Header:** 'Is there a process to conduct ongoing monitoring to identify any adverse impacts to the performance and security of the AI functionality, as well as to privacy, civil rights, and civil liberties? Practice: Conduct Ongoing Monitoring for Performance and Potential Adverse Impacts' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether a process to conduct ongoing monitoring for the AI use case has been established, consistent with Section 4(b)(iii) of the appendix to OMB Memorandum M-25-21. Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Yes, sufficient monitoring protocols have been established | | b) Development of monitoring protocols is in-progess | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Yes, sufficient monitoring protocols have been established | **Yes - Monitoring Established** | | b) Development of monitoring protocols is in-progress | **In-progress** | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | --- ### 33) `hi_training_established` - **Original Excel Header:** 'Has the agency established sufficient and periodic training for operators of the AI to interpret and act on the its output and managed associated risks? Practice: Ensure Adequate Human Training and Assessment' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether sufficient and periodic training for operators of the AI use case has been established, consistent with Section 4(b)(iv) of the appendix to OMB Memorandum M-25-21. Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Yes, sufficient and periodic training has been established | | b) Establishment of sufficient and periodic training is in-progress | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Yes, sufficient and periodic training has been established | **Training Established** | | b) Establishment of sufficient and periodic training is in-progress | **Training In-progress** | | c) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | --- ### 34) `hi_failsafe_presence` - **Original Excel Header:** 'Does this AI use case have an appropriate fail-safe that minimizes the risk of significant harm? Practice: Provide Additional Human Oversight, Intervention, and Accountability' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether the AI use case has an appropriate fail-safe that minimizes the risk of significant harm, consistent with Section 4(b)(v) of the appendix to OMB Memorandum M-25-21. Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Yes | | b) Not applicable | | c) In-progress | | d) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Yes | **Yes** | | b) Not applicable | **Not Applicable** | | c) In-progress | **In-progress** | | d) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | --- ### 35) `hi_appeal_process` - **Original Excel Header:** 'Is there an established appeal process in the event that an impacted individual would like to appeal or contest the AI system’s outcome? Practice: Offer Consistent Remedies or Appeals' - **Data Type:** `multiple_choice` - **Instructions:** 'Whether there is an established appeal process for this AI use case in the event that an impacted individual would like to appeal or contest the AI system’s outcome, consistent with Section 4(b)(vi) of the appendix to OMB Memorandum M-25-21. Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Yes, an appropriate appeal process has been established | | b) Not applicable | | c) Establishment of an appropriate appeal process is in-progress | | d) Law, operational limitations, or governmentwide guidance precludes an opportunity for an individual to appeal | | e) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Yes, an appropriate appeal process has been established | **Appeal Process Established** | | b) Not applicable | **Not Applicable** | | c) Establishment of an appropriate appeal process is in-progress | **Appeal Process In-progress** | | d) Law, operational limitations, or governmentwide guidance precludes an opportunity for an individual to appeal | **Appeal Precluded by Law** | | e) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | --- ### 36) `hi_public_consultation` - **Original Excel Header:** 'What steps has the agency taken to consult and incorporate feedback from end users of this AI use case and the public? Practice: Consult and Incorporate Feedback from End Users and the Public' - **Data Type:** `select_multiple` - **Instructions:** 'Identification of what steps the agency has taken to consult and incorporate feedback from end users of this AI use case and the public, consistent with Section 4(b)(vii) of the appendix to OMB Memorandum M-25-21. Required: Yes, only for high-impact deployed use cases' **Valid Selections:** | Valid Value | | :--- | | a) Direct usability testing | | b) General solicitations of feedback and comments from the public | | c) Public hearings or meetings | | d) Other | | e) In-progress | | f) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **Shorthand Recoding Map:** | Original Value | Shorthand Code | | :--- | :--- | | a) Direct usability testing | **Usability Testing** | | b) General solicitations of feedback and comments from the public | **Public Feedback Solicitations** | | c) Public hearings or meetings | **Public Hearings** | | d) Other | **Other** | | e) In-progress | **In-progress** | | f) Agency CAIO has waived this minimum practice and reported such waiver to OMB | **CAIO Waived** | ---