Foi Systems With Ai Agents

North Mini Code. Cohere's first model for developers.

Learn more

Aug 08, 2025

6 minutes read

Modernizing freedom of information (FOI) request systems with AI

Freedom of information (FOI) request systems need support. The manual, rule-based nature of the FOI process makes them a prime candidate for AI automation in the public sector.

!Blog Post Featured Image

_Updated: _

Freedom of information (FOI) requests are a vital pillar of democratic public administration, giving citizens, journalists, researchers and civil society organizations a formal channel to request government-held information.

But governments worldwide are finding these public records requests increasingly difficult to fulfill. For example, the US government received more than 1.7 million FOIA requests in FY 2025, while backlogs across major federal agencies rose nearly 27% as full-time FOIA staffing fell 16%.

These backlogs strain public resources, frustrate citizens, and weaken a core mechanism of government transparency. This is where AI can help. Processing FOI requests involves interpreting natural-language submissions, searching fragmented systems, classifying documents, identifying sensitive material, and preparing records for release — all tasks where AI can support faster, more consistent workflows.

In particular, AI agents can be used to enhance the agility and responsiveness of FOI systems, helping reduce processing times, enabling record officers to handle more cases, and lowering error rates. Agentic platforms, like North, can be deployed privately and securely, addressing a key concern for agencies adopting AI in public administration.

Governments we work with are actively exploring the use of AI to enhance nearly every aspect of the current FOI process, from triaging requests to assessing their sensitivity and completeness, to retrieving records.

##### Current FOI bottlenecks in public administration

The basic premise of FOI laws is simple: If you ask to see a government record, the government has to either show it to you or give a legally valid reason for keeping it secret.

Unfortunately, the reality of public access often falls short of this standard. Backlogs and delays are an issue in many jurisdictions that maintain FOI or public records laws, and the slow pace of access blunts their effectiveness as tools for accountability.

There are many reasons for this slow pace, but understaffing and reliance on manual processes are two of the most persistent constraints. Faced with competing demands and limited resources, agencies often struggle to prioritize FOI, leaving requesters waiting longer for access to public information.

###### How long does it take to process a FOIA request?

In the US, the average processing time for simple-track FOIA requests was 44 days in FY 2024, up from 39.4 days in FY 2023.

Complex requests typically take longer since they often involve larger record volumes, searches across multiple locations, or consultations with other agencies. In FY 2024, only 80.66% of complex requests were processed within 100 days.

###### How the process currently works (without AI)

Most FOI systems start with a portal where members of the public cansubmit FOI or public records requests. While some countries have a single portal, others, such as the United States, maintain separate intake systems for each agency, resulting in fragmented intake processes and uneven routing between record offices. Even where systems are unified on the front end, each department typically manages its own records and maintains its own public records officer for handling requests.

From there, each request is assigned to a specific officer who communicates directly with the requester. Often, there are particular notification requirements (e.g., when a request is received or when a certain amount of time has passed), but communication is typically handled manually through email or phone calls. The officer may sometimes propose modifications to the request, narrowing it in ways that result in faster fulfillment or fewer extraneous records; however, these modifications must be approved by the requesting party in advance.

The biggest pain point is redaction. Once relevant documents have been assembled, records officers must review them carefully to remove trade secrets, classified information, or other sensitive material. Since this review is typically performed manually, requests that span hundreds of pages can become highly labor-intensive.

##### How AI can transform the FOI process

The good news for governments struggling to meet public transparency expectations is that FOI workflows align closely with AI’s strengths.

A significant portion of handling each FOI request involves repetitive, rule-based tasks, such as reviewing emails, memos, and reports to identify relevant documents and verify them against statutory exemptions. But even when the rules are clear, applying them can be difficult because relevant information is rarely stored in one neat, searchable place.

For example, one common FOI request asks for communications involving a specific office holder during a defined period, related to particular keywords or subject matter. While the phrasing of the request may be simple, it could involve collating records from half a dozen different systems over the course of years. It’s the kind of request that can be frustrating, tedious, and time-consuming to perform manually.

AI can help with this early discovery work by translating broad written requests into structured search logic, applying that logic across fragmented systems, and giving records officers a focused set of relevant materials to review. Models can also be trained or tuned on historical requests, agency guidance, and past FOI review decisions. This allows them to better...