# AI for FOIA and Public Records Requests

> Government · AI Course · GOV-11
> Source: https://ibl.ai/solutions/government/course/ai-for-foia-and-public-records-requests
> Last updated: 2026-11-18

**Use AI agents to log, search, de-duplicate and organize records for FOIA and state public records requests, while records officers make every exemption and release decision.**

## The Short Answer

**AI can speed FOIA and public records responses by logging and acknowledging requests, searching record systems, de-duplicating results and flagging possible exemptions for review, while a records officer makes every withholding decision. ibl.ai runs this on agency infrastructure, on-premise or air-gapped, and you own all the code and the data, so requested records never leave agency control.**

On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network.

[Request Access](https://ibl.ai/contact) · [Explore Government](https://ibl.ai/solutions/government)

## Course facts

- **Level:** Intermediate
- **Duration:** 6 hours across 8 modules
- **Format:** Cohort workshop with a hands-on lab
- **Modules:** 8
- **Catalog code:** GOV-11
- **Frameworks covered:** FOIA (5 U.S.C. § 552), State public records laws, NARA records management, NIST AI RMF

## What is this course about?

Records requests keep rising while staff and statutory deadlines do not move. This course applies AI to the slow parts of a request, intake, search, de-duplication and review preparation, and keeps the legal decisions with records officers: what an agent may flag, what it must never decide, and how to keep a defensible record of review.

## Who is this course for?

- FOIA officers and public records coordinators
- Agency counsel who review withholdings
- Records managers
- Agency IT staff who support search and redaction tools

### What do I need before starting?

- Working knowledge of your agency's records request process
- Recommended: Records, FOIA, and AI (GOV-5)

## What will I be able to do afterwards?

- Measure where time goes across the request lifecycle
- Design request intake and routing an agent can run
- Document a search across record systems that the agency can defend
- Configure exemption flagging that leaves every decision to a person
- Keep a review record that supports the release decision

## What does each module cover?

### Module 1 — Where does the time go in a records request?

Intake, search, review, redaction and release measured against the statutory clock; federal FOIA sets 20 working days for a determination, and state laws differ. _(45 min)_

**Objectives**

- Map the request lifecycle
- Measure time by stage
- Identify the stages AI can shorten

**Topics:** Request lifecycle · Statutory deadlines · Backlog analysis · State law variation

**Activity:** Measure stage times from a sample of closed requests.

### Module 2 — How do you triage and route an incoming request?

Logging, acknowledgment, scope clarification and tracks for simple, complex and expedited requests. _(45 min)_

**Objectives**

- Classify requests by complexity
- Draft acknowledgment and clarification letters for review
- Route requests to record custodians

**Topics:** Intake and logging · Scope clarification · Processing tracks · Expedited processing

**Activity:** Have an agent classify twenty historical requests and compare with the actual handling.

### Module 3 — How does an agent search across record systems?

Email, shared drives and case systems, with the search terms and locations documented. _(45 min)_

**Objectives**

- Translate a request into documented search terms
- Search across systems with consistent methods
- Record the search for later review

**Topics:** Search term design · Cross-system search · Search documentation · Adequacy of search

**Activity:** Run and document a search for one historical request.

### Module 4 — How do you de-duplicate and organize a responsive set?

Threads, attachments and near-duplicates collapse a set to what a reviewer must read. _(45 min)_

**Objectives**

- De-duplicate exact and near-duplicate records
- Group email threads and attachments
- Order the set for review

**Topics:** Exact duplicates · Near-duplicates · Email threading · Review ordering

**Activity:** De-duplicate a responsive set and measure the reduction.

### Module 5 — How should an agent flag exemptions without deciding them?

Flags carry a reason and a passage; the withholding decision, including whether disclosure would cause foreseeable harm, stays with a person. _(45 min)_

**Objectives**

- Configure exemption flags with stated reasons
- Measure flag precision and recall
- Keep every decision with a records officer

**Topics:** Exemption categories · Flag reasons · Foreseeable harm · Human determination

**Activity:** Compare agent flags with a records officer's decisions on a closed request.

### Module 6 — How do you redact and keep a defensible record of review?

Redaction, withholding logs and an audit trail that supports the release letter and any appeal. _(45 min)_

**Objectives**

- Apply redactions that cannot be reversed
- Log every withholding with its basis
- Keep an audit trail of review

**Topics:** Redaction · Withholding logs · Audit trail · Appeals

**Activity:** Produce a withholding log for a reviewed set.

### Module 7 — How do you communicate with requesters in plain language?

Status updates, estimated completion dates and proactive disclosure of frequently requested records. _(45 min)_

**Objectives**

- Draft status updates for review
- Identify records to publish proactively
- Write release letters in plain language

**Topics:** Status communication · Estimated completion dates · Proactive disclosure · Plain language

**Activity:** Rewrite a release letter in plain language and list five records to publish proactively.

### Module 8 — Building the request-processing workflow

The lab: an end-to-end workflow on a closed request, from intake to release package. _(45 min)_

**Objectives**

- Configure intake, search, de-duplication and flagging
- Run a review with a records officer
- Measure time saved against the baseline

**Topics:** Workflow configuration · Human review · Time measurement · Pilot plan

**Activity:** Process a closed request end to end and compare with the original handling time.

## What is the capstone project?

**Records request workflow and pilot plan.** Design an AI-assisted request workflow for your agency: intake and routing, documented search, de-duplication, exemption flagging with human decisions, a withholding log and requester communications, with a measured baseline.

_Deliverable:_ A workflow design, a processed sample request with its withholding log, and a pilot plan approved by the agency's FOIA officer.

## How are learners assessed?

- Search documentation reviewed for completeness by a records officer
- Exemption flags compared with human decisions on a closed request
- Withholding log checked for a stated basis on every entry

## What ships with the course?

- **Facilitator guide.** Session-by-session running order, discussion prompts, and the questions that reliably derail a room.
- **Learner workbook.** Exercises, checklists, and the templates each module's activity produces.
- **Hands-on lab environment.** A sandboxed ibl.ai deployment so exercises run against real agents, not screenshots.
- **Assessment bank.** Scenario questions and rubric criteria mapped to each stated learning outcome.
- **Source bibliography.** Every primary regulation and standard cited on this page, linked and dated.

## Which AI agents does this course use?

- [Knowledge Agent](https://ibl.ai/solutions/government/agent/knowledge-agent)
- [Compliance Agent](https://ibl.ai/solutions/government/agent/compliance-agent)
- [Constituent Communication Agent](https://ibl.ai/solutions/government/agent/constituent-communication-agent)
- [Citizen Services Agent](https://ibl.ai/solutions/government/agent/citizen-services-agent)

## Where does the course material come from?

Every module is grounded in primary sources — the regulation, standard, or research itself, not a summary of it. Each was resolved at authoring time.

- [5 U.S. Code § 552: Public information; agency rules, opinions, orders, records, and proceedings](https://www.law.cornell.edu/uscode/text/5/552) — Cornell Legal Information Institute. The FOIA statute: response deadlines, exemptions and the foreseeable harm standard.
- [Department of Justice Guide to the Freedom of Information Act](https://www.justice.gov/oip/doj-guide-freedom-information-act-0) — U.S. Department of Justice, Office of Information Policy. Chapter-level guidance on search, exemptions and processing.
- [OIP Guidance](https://www.justice.gov/oip/oip-guidance) — U.S. Department of Justice, Office of Information Policy. Guidance on proactive disclosure, communication with requesters and processing practice.
- [FOIA.gov](https://www.foia.gov/) — U.S. Department of Justice. Request handling and reporting context across federal agencies.
- [Federal Records Management](https://www.archives.gov/records-mgmt) — National Archives. Records management requirements for the record systems an agent searches.
- [AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) — NIST. The govern, map, measure and manage functions used to structure risk review.

## Delivery notes

Binding guidance for anyone preparing and delivering this course:

- State public records laws differ in deadlines, exemptions and fees. Deliver module 1 with the learner's own state statute alongside the federal one.
- Module 5 is the line the course exists to hold: an agent flags, a person decides. Any demonstration that blurs it undermines the course.
- Use closed requests for every lab. Never run a live request through a pilot before counsel approves the workflow.
- Coordinate with GOV-5, which covers AI-generated material as records; this course covers AI as a tool for processing requests.

## Why run AI training on a platform you own?

- **You own the course, not a licence to it.** Course content, learner data, and the platform run inside your perimeter — you own all the code and the data.
- **Model-agnostic delivery.** Run the course's AI components on any LLM — Claude, GPT, Llama, Gemini, Command — and switch anytime.
- **No per-seat training licences.** Usage-based or self-hosted, so cost tracks actual use rather than headcount.
- **Deploy anywhere.** Cloud, private VPC, on-premise, or fully air-gapped — including for cohorts that cannot use public AI tools.

## Frequently asked questions

### What does the AI for FOIA and Public Records Requests course cover?

Records requests keep rising while staff and statutory deadlines do not move. This course applies AI to the slow parts of a request, intake, search, de-duplication and review preparation, and keeps the legal decisions with records officers: what an agent may flag, what it must never decide, and how to keep a defensible record of review. It runs 6 hours across 8 modules, at intermediate level, and closes with a capstone: Records request workflow and pilot plan.

### Who should take AI for FOIA and Public Records Requests?

It is written for FOIA officers and public records coordinators, Agency counsel who review withholdings, Records managers, Agency IT staff who support search and redaction tools. Prerequisites: Working knowledge of your agency's records request process; Recommended: Records, FOIA, and AI (GOV-5).

### Can we run this course on our own infrastructure?

Yes. ibl.ai is model-agnostic and deploy-anywhere — cloud, private VPC, on-premise, or fully air-gapped — and you own all the code and the data. Cohort data, submissions, and any material learners upload stay inside your perimeter, which matters for government teams that cannot send work to a public AI tool.

### How do we get access to AI for FOIA and Public Records Requests?

Request access and we will set it up for your cohort — hosted by ibl.ai, or running against your own deployment. Tell us the group size and timing you need, and whether it should run inside your own perimeter.

### How much does AI training for government cost on ibl.ai?

There is no per-seat pricing — you pay for usage or self-host and pay only for the infrastructure, so a 5,000-person rollout does not cost 5,000 licences. 1.6M+ users across 400+ organizations run the platform this way, including NVIDIA, MIT, and Syracuse University.

## More Government courses

- [Getting an AI System Authorized: FedRAMP and NIST 800-53](https://ibl.ai/solutions/government/course/authorizing-ai-fedramp-nist-800-53): The authorization path for AI in a federal or state agency — control selection, boundary definition, and why an LLM complicates the system security plan.
- [Sovereign and Air-Gapped AI for Public Agencies](https://ibl.ai/solutions/government/course/sovereign-air-gapped-ai-public-agencies): Run capable AI with no internet egress — model selection, air-gapped update paths, and the operational realities of a disconnected deployment.
- [Citizen Service Agents: Design, Escalation, and Accessibility](https://ibl.ai/solutions/government/course/citizen-service-agents): Public-facing AI where the user has no alternative provider — plain language, Section 508 conformance, language access, and escalation that never traps a constituent.
- [AI in Public Procurement: Writing an RFP That Gets Real Bids](https://ibl.ai/solutions/government/course/ai-in-public-procurement): Specify AI in a solicitation so you get comparable, honest proposals — required disclosures, evaluation criteria, and contract terms that preserve agency control.
- [Records, FOIA, and AI: Retention When an Agent Writes](https://ibl.ai/solutions/government/course/records-foia-and-ai): What happens to public records law when an AI drafts the memo — retention, prompt logs as records, and responding to a request that reaches an AI system.
- [AI Governance Inside a Public Agency](https://ibl.ai/solutions/government/course/ai-governance-public-agency): Stand up an agency AI governance program — inventory, use-case review board, impact assessment, and public transparency reporting.
