IE11 Not Supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

Miami Turns to AI to Produce Zoning Verification Letters

The city will work with C3.ai as part of a three-month pilot program to improve and expedite its production of zoning verification letters through artificial intelligence.

Closeup of a person writing in a notebook overlayed with the word "AI." White background.
Shutterstock
The city of Miami has partnered with C3.ai to carry out a three-month pilot program to improve and expedite its production of zoning verification letters through AI.

The Miami City Commission has accepted a $250,000 donation from Google for a three-month subscription to the C3 Generative AI for Government Program to develop generative artificial intelligence in a Google Cloud environment to assist with producing zoning verification letters for T5 transect zones.

The T5 transect zone in Miami is classified as an urban center with higher-density, mixed-use buildings comprising retail, office and apartment units.

In this case, Google’s donation will allow the city “to explore the use of AI in producing zoning verification letters more efficiently while evaluating its broader application in city services.”

The city will pay a nominal one-time fee of $1 for the duration of the pilot program.

City documents state that the pilot “will be limited to 20GB of publicly available, unstructured documents (up to a total of 100,000 pages) and structured datasets provided by the city of Miami, and up to twenty (20) internal city of Miami users.”

Personally identifiable information will be excluded from the scope of this pilot, and Google Gemini will be used as the large language model for the project.

Specific objectives for the project include:
  • Loading in-scope data into the C3 Generative AI for Government Program's product in C3’s Google Cloud environment 
  • Tuning underlying AI retrieval models to reflect the knowledge contained in source documents and configuring Gemini to provide relevant answers while minimizing hallucinations 
  • Configuring the AI to generate parcel-specific reports 
  • Collecting feedback from city of Miami users to enable ongoing model tuning and improvements 
  • Conducting integration, quality assurance, performance and user-acceptance testing 
More information about the pilot program can be found online.
Katya Diaz is an Orlando-based e.Republic staff writer. She has a bachelor’s degree in journalism and a master’s degree in global strategic communications from Florida International University.