The use of artificial intelligence platforms is severely limited under a policy the City of Pittsburgh released to PublicSource in response to a Right-to-Know Law request.
Drafted in October and updated in February, the city’s policy on the use of generative AI — computer systems that create new content — bars city staff from including private city data in interactions with tools like ChatGPT and Bing Chat.
Also prohibited is use of AI in any applications that impact the rights or safety of residents.
“Don’t ask Generative AI for knowledge,” the policy instructs, nor for decisions, incident reports or generation of images or video.
When AI is used, city staff are to “mind the bias” that can be deep in the code “based on past stereotypes.” And all use of AI must be disclosed to any audiences that receive the end product, plus logged and tracked.
The city released the policy after PublicSource published a story describing the emerging approach to AI in local government. Allegheny County has passed use of the platforms while a working group develops a policy.
Government use of generative AI comes with risks, including the possibility of convincing fake images, that could erode public trust. Experts worry that officials haven’t properly regulated those algorithmic tools that have been around for years.
Rich Lord is the managing editor at PublicSource and can be reached at rich@publicsource.org.




