Attention: You are using an outdated browser, device or you do not have the latest version of JavaScript downloaded and so this website may not work as expected. Please download the latest software or switch device to avoid further issues.
| 9 Jul 2026 | |
| Written by J.B. Wogan | |
| Press Releases and Statements |
WASHINGTON, D.C., July 9, 2026 — The Data Foundation today launched Evi, an AI assistant built to help users search and analyze the Evidence Act Hub, a public archive of federal evidence-building documents at evidenceact.org. Currently in a beta phase, Evi answers questions about agency Learning Agendas, Annual Evaluation Plans, and Capacity Assessments, and cites the source document for every answer. Evi was co-developed with Data Foundation Senior Fellow Ted Kaouk, former Chair of the Federal Chief Data Officers Council, using Structured Notes, an AI platform built by his firm, Generative Work Labs.
Under the Foundations for Evidence-Based Policymaking Act of 2018 (the Evidence Act), federal agencies have produced hundreds of documents, such as learning agendas, evaluation plans, and capacity assessments, to describe how they plan to build and use evidence. Until now, understanding this material meant manually downloading and reading dozens of separate PDFs from different agency websites, each formatted differently. A simple cross-agency question — for example, "Which agencies are researching fraud?" — could take days of manual review to answer, but with Evi it can be answered in seconds.
"The enduring promise of the Evidence Act has always been about building and sustaining a culture of evidence-informed decision-making," said Nick Hart, President and CEO of the Data Foundation. "For too long, valuable insights from agency learning agendas and evaluation plans have been trapped in siloed PDFs, making it nearly impossible to see the full picture of the federal government’s research needs. Adding an AI layer to the Evidence Hub, which we call Evi, moves us from merely monitoring compliance with this important law to now promoting the usefulness of the tools as Congress intended, making complex federal resources searchable and actionable."
The Center for Evidence Capacity (CEC), a Data Foundation program, launched the Evidence Act Hub in December 2025. The Hub is a centralized public archive documenting seven years of Evidence Act implementation across the federal government. It enables access to agency learning agendas, evaluation plans, capacity assessments, and related strategic planning documents in one place.
With the retirement of evaluation.gov and cdo.gov, as well as the migration of public information on the Evidence Act to councils.gov, the Evidence Act Hub now serves as the principal, non-federal public repository for this material. It ensures the documents remain accessible even if individual agencies remove or restructure their own web content, for example, during a change in leadership or a website redesign.
Evi is an agentic artificial intelligence query layer built on top of the Hub's existing document collections. In natural language, users can ask Evi to:
In beta testing, Evi correctly identified agencies with fraud-related learning agenda questions and, in a follow-up query, correctly linked the concept of "integrity" to fraud prevention, a task that previously required extensive manual review across agency documents.
Every Evi response is grounded in documents housed in the Hub and includes a citation to the original source, so users can verify the answer and read further.
The Data Foundation built Evi and the Evidence Act Hub to address four specific problems:
The Data Foundation is a Washington, DC-based, non-profit, non-partisan organization. It is a trusted authority on the use of open, accessible data to fuel a more efficient, effective, and accountable government; spark innovation; and provide insights to the country's most pressing challenges. It conducts research, facilitates collaborative thought leadership, and promotes advocacy programs that advance practical policies for the creation and use of accessible, trustworthy data and evidence. The Data Foundation is recognized by Candid Guidestar with the Platinum Seal of Transparency and by Charity Navigator as a 4-Star non-profit. To learn more, visit www.datafoundation.org. (LEI: 254900I43CTC59RFW495)
What is Evi?
Evi is an AI assistant built by the Data Foundation to help users search and analyze documents in the Evidence Act Hub. Currently in beta, Evi answers questions about federal agency learning agendas, evaluation plans, and capacity assessments, with citations to the original source for every answer.
Who built Evi?
Evi was co-developed by the Data Foundation, Data Foundation Senior Fellow Ted Kaouk, and his firm generativework.ai. Its platform, Structured Notes, turns unstructured documents into AI-ready intelligence, by building knowledge graphs that surface hidden insights and enable information retrieval and workflow execution with AI. Learn more at structurednotes.app.
What can Evi help me with?
Evi can help you explore the Evidence Act Hub’s document repository by answering natural-language questions. For example, you can ask Evi to compare evidence-building priorities across agencies, summarize learning agendas for a specific department, or locate key policy documents like OMB guidance.
How does Evi generate answers?
Evi is "grounded," meaning the assistant answers questions based on the verified documents provided within the Evidence Act Hub. Every answer includes inline citations to the specific document and section being referenced, so you can always check the original source yourself.
How is Evi different from a general AI chatbot?
Evi answers based on documents housed in the Evidence Act Hub and provides an inline citation to the source document with every answer, rather than drawing on general knowledge.
Can Evi make mistakes?
Yes. Evi is an AI assistant, and especially while in beta, it can occasionally misread a document or summarize one imperfectly. If you come across an answer that looks inaccurate, we'd genuinely welcome hearing about it. Please reach out to the Data Foundation team at evidencehub@datafoundation.org.
Can Evi give me legal or policy advice?
No. Evi is strictly an informational tool. The assistant cannot provide legal advice, policy advocacy, or opinions on whether a specific law or policy is "good" or "bad." Evi also will not speculate on agency intent beyond what is explicitly documented in the Hub’s materials.
What if Evi doesn't know the answer to my question?
Because Evi is designed to be highly reliable and accurate, the assistant is programmed to be honest about its limits. If a question falls outside the scope of our materials, or if the Hub’s documents do not contain the answer, Evi will politely explain that and suggest that you reach out to the Data Foundation staff at evidencehub@datafoundation.org for further assistance.
What is the Evidence Act Hub?
The Evidence Act Hub is a public archive at evidenceact.org, launched in December 2025 by the Center for Evidence Capacity, a Data Foundation program. It centralizes federal documents required under the Foundations for Evidence-Based Policymaking Act of 2018.
Why was the Evidence Act Hub created?
Federal evidence-building documents were previously scattered across dozens of individual agency websites in inconsistent formats, making cross-agency analysis slow and manual. The Hub centralizes these documents in one searchable location, and now serves as the principal public archive following the retirement of evaluation.gov.
How can I access Evi?
Evi is available at evidenceact.org.
J.B. Wogan
Director of Communications, Data Foundation
Phone: (202) 964-1130
Email: media@datafoundation.org
