The OECD.AI Policy Navigator

Our policy navigator is a living repository from more than 80 jurisdictions and organisations. Use the filters to browse initiatives and find what you are looking for.

Ulysses - Bill Procedural Routing Analysis


Added by:   OECD analyst
Added on:   05 Aug 2026
Updated by:   OECD analyst
Updated on:   05 Aug 2026

Ulysses - Bill Procedural Routing Analysis is an AI system incoming legislative bills to identify procedural requirements and committee assignments, then produces a dispatch draft suggesting the bill's processing regime, committee routing order, and consideration procedures. It supports the Clerk’s staff before sending the bill to be discussed in the legislative committees. The Clerk staff are required to make many deep analyses of each incoming bill, which requires a lot of time.

Name in original language

Ulysses - Análise de Roteamento Procedimental de Propostas Legislativas

Initiative overview

The process of distributing legislative bills to the thematic committees – Initial Dispatch – requires manual and technical analysis by the Clerk staff. This activity, although critical for the proper progress of legislative processing, is susceptible to inconsistencies, delays, and work overload. Additionally, the growing volume of bills makes the process increasingly challenging in terms of time and accuracy. 

That process is composed of five modules and implemented by AI agents that evaluate the legislative document from different perspectives:

  • Assessment of the bill's procedural and constitutional requirements
  • Search for similar legislative documents already being processed in the Brazilian Chamber of Deputies, so that, instead of having independent processing, they can be attached to a similar bill
  • Definition of the form of analysis of the legislative matter, that is, whether by characteristic or rule it will first be evaluated by thematic committees before being voted on in plenary, or whether it is possible for the legislative proposal to become law as long as it has a conclusive opinion in the thematic committees through which it must pass. 

Below are the alternatives for form of bill analysis procedure:

  • Bill subject to conclusive discussion by committees
  • Bill subject to plenary discussion,
  • Bill subject to internal discussions in committees.
  • Definition of the processing regime: under Brazilian legislation, legislative matters can be processed under the following regimes: Ordinary, Priority, Urgency, Special.
  • Distribution to thematic committees for evaluation of merit, financial adequacy, and constitutionality. 

In the Chamber of Deputies of Brazil there are 30 permanent committees. All these analyses approaches are supported by AI agents to choose the best option, complying with the procedural and constitutional rights and rules involved.

Other relevant details

Results, outcomes and impacts: 1) Increase operational efficiency, 2) Reduction of the initial processing time for bills, contributing to greater speed in the legislative process as a whole, 3) Greater accuracy in the distribution of bills, based on consistent technical criteria, 4) Support for decision-making by technical teams, who will be able to rely on automated recommendations while remaining in the role of supervisors and validators of the process, 5) Standardisation of actions, 6) Process scalability, allowing the volume of bills to keep pace with growing demand without compromising quality or distribution deadlines. Challenges and lessons learned: The greatest challenge was to provide a scalable infrastructure with adequate processing capacity to respond effectively and efficiently. Contracting a platform that aggregated multiple AI models from different providers in one place was crucial to the success of this initiative. As a prerequisite for the application's success: strategic alignment with business objectives; availability of different versions of AI models; feasibility of integration with legacy systems; multidisciplinary team; continuous user feedback for iterative improvement.

About the policy initiative


Category:

  • AI policy initiatives, programmes and projects

Initiative type:

  • AI use cases/projects in the public sector

Status:

  • Active

Start Year:

  • 2025

Target Sectors:


OECD AI Principles:


Other relevant urls: